The dates displayed for an article provide information on when various publication milestones were reached at the journal that has published the article
activities on preceding journals at which the article was previously under consideration are not shown (for instance submission
All content on this site: Copyright © 2025 Elsevier B.V.
The Simons Foundation is pleased to announce the latest recipients of its <a href="/grant/travel-support-for-mathematicians/?tab=rfa">Travel Support for Mathematicians Award</a>
the Simons Foundation’s Mathematics and Physical Sciences division invites applications for its Travel Support for Mathematicians program
which is intended to stimulate collaboration in the field of mathematics primarily through the funding of travel and related expenditures
Each awardee receives up to $6,000 per year for travel and research expenses such as scientific travel by the PI
An additional $1,000 per year is allocated towards department discretionary funds to enhance the department&#8217;s research atmosphere
The Simons Foundation is awarding $5.7 million total in grants to 140 grantees this year
The approach of archeology to society has changed in Spain in recent years
evidenced the need to involve local communities in archaeology
opening up to new proposals on the political use of this discipline
The integration of local communities with heritage was valued
actively reinforcing their identity through History
this example has reached Spain thanks to the promotion of studies by professionals
Social participation can be established in two main axes: actively
laboratory work and dissemination or passively
but always seeking understanding in this participation
the conservation and valuation of archaeological heritage.&nbsp;
The interaction of the different work proposals for the practice of archeology has had a positive impact on social articulation
promoting conservation and dissemination measures not only inherent to local organizations
An example is the El Ponderal cultural association in Hoyo de Manzanares
made up of volunteers motivated by research and promotion of the history of their municipality
after having actively participated in the excavation of the La Cabilda site
is the “Heritage in the hands of young people” project
in coordination with the management of the Los Abetos School
in which a large group of 3rd and 4th ESO students developed a theoretical and archaeological practice in different actions in its environment or the work carried out at the Marqués de Santillana High School Institute in Colmenar Viejo
(Team A of Archeology) (2014):&nbsp;Two mining-metallurgical enclaves during Late Antiquity in the center of the peninsula: Navalvillar and Navalahija (Colmenar Viejo
The primitive castle of Real de Manzanares
(Team A of Archeology) (2013):&nbsp;In search of the lost magnetite
Iron metallurgy and village organization during late Antiquity in Navalvillar and Navalahija (Colmenar Viejo
Proceedings of the X Conference on Archaeological Heritage in the Community of Madrid
(Team A of Archeology) (2014):&nbsp;Settlement during Late Antiquity and the Middle Ages in the Madrid presierra: Cuenca Alta del Manzanares
(Archeology Team A) (2016):&nbsp;The exploitation of iron in Late Antiquity in the upper Manzanares Basin
Historic Mining and Metallurgy in Southwest Europe
Proceedings of the IX International Congress on Historic Mining and Metallurgy in Southwest Europe
MA (CCHS-CSIC) (In press):&nbsp;Glass of Late Antiquity in the Upper Manzanares Basin
VII Conference on Archeology in the Duero Valley
(2005):&nbsp;Guide to the archaeological site of Remedios
A rural cemetery during Late Antiquity (1th century AD)
historical and artistic heritage of Colmenar Viejo nº XNUMX
(1967):&nbsp;The castles of Manzanares el Real
Annals of the Institute of Madrilenian Studies
(2005):&nbsp;The first castle of Manzanares el Real ?
(Archeology Team A) (2016):&nbsp;The archaeological site of La Cabilda (Hoyo de Manzanares)
(Archeology Team A) and GIMENO (CIL) (2016):&nbsp;Two latentiguan rings with inscription in the mountains of Madrid
Sylloge Epigraphica Barcinonensis (SEBarc) XIV
(Archeology Team A) (2016):&nbsp;The late medieval site of La Cabilda
(2009):&nbsp;History and guide of the mines
From the bowels of the earth: Guide to the mines and quarries of Colmenar Viejo
Historical and Artistic Heritage of Colmenar Viejo
(2015):&nbsp;Visigoth landscape in the Upper Manzanares Basin (Sierra de Guadarrama): archaeological analysis of the Navalvillar site (Colmenar Viejo
Archeology and Prehistory of the Interior Peninsular
(Team A of Archeology) (2015):&nbsp;Iron in the Navalvillar and Navalahija deposits during Late Antiquity
fortifications and walled enclosures in the Community of Madrid
General Directorate of Historical Heritage
<a href="/articles/s41598-024-62061-8/metrics" data-track="click" data-track-action="view metrics" data-track-label="link" rel="nofollow">Metrics details</a>
Glioblastoma (GB) is a devastating tumor of the central nervous system characterized by a poor prognosis
One of the best-established predictive biomarker in IDH-wildtype GB is O6-methylguanine-DNA methyltransferase (MGMT) methylation (mMGMT)
which is associated with improved treatment response and survival
current efforts to monitor GB patients through mMGMT detection have proven unsuccessful
Small extracellular vesicles (sEVs) hold potential as a key element that could revolutionize clinical practice by offering new possibilities for liquid biopsy
This study aimed to determine the utility of sEV-based liquid biopsy as a predictive biomarker and disease monitoring tool in patients with IDH-wildtype GB
Our findings show consistent results with tissue-based analysis
achieving a remarkable sensitivity of 85.7% for detecting mMGMT in liquid biopsy
we suggested that liquid biopsy assessment of sEV-DNA could be a powerful tool for monitoring disease progression in IDH-wildtype GB patients
This study highlights the critical significance of overcoming molecular underdetection
which can lead to missed treatment opportunities and misdiagnoses
possibly resulting in ineffective therapies
The outcomes of our research significantly contribute to the field of sEV-DNA-based liquid biopsy
providing valuable insights into tumor tissue heterogeneity and establishing it as a promising tool for detecting GB biomarkers
These results have substantial implications for advancing predictive and therapeutic approaches in the context of GB and warrant further exploration and validation in clinical settings
this study aimed to determine the utility of sEV-based liquid biopsy to analyze mMGMT status as a predictive biomarker and disease monitoring tool in patients with glioblastoma
Surgical complete resection was defined based on absence of tumor contrast enhancement on early postoperative MRI (&lt; 72 h from surgery)
This study was approved by the institutional Ethics Committee (PI-2887) according to the Helsinki Declaration
Blood samples were obtained in EDTAK2 BD Vacutainer tubes (Becton Dickinson
USA) after surgery and before starting the concomitant treatment with STUPP protocol
Plasma was obtained by a first centrifugation at 2500g for 10 min at 4 °C
The supernatant was transferred to a new tube followed by a second step of centrifugation at 35,000g for 20 min at 4 °C
Plasma samples were stored in 1.5 ml aliquots at − 80 °C until further processing
the PBS-diluted plasma was filled into 13.5 ml thick-walled polycarbonate reusable tubes specific for UC processing (Beckman Coulter
Samples were ultracentrifuged at 100,000×g for 70 min at 4 °C using a fixed-angle rotor (Rotor 70.1 Ti
K factor of 36) in an Optima L-100 XP ultracentrifuge (Beckman Coulter
The pellet was washed with 7 ml of filtrated PBS and ultracentrifuged again under the same conditions
The pellet containing the sEVs was suspended in 150 μl of filtered PBS
UK) employs NTA technology that enables precise analysis of the size distribution and concentration of all types of suspended nanoparticles
From 10 μl of the sEV suspension obtained after UC
a 1:100 dilution was carried out with filtered PBS to achieve the concentration range recommended by the manufacturer (108–109 particles/ml)
This dilution was injected into the device
and two 60-s videos for each sample were generated and analyzed by NTA 3.0 software
providing particle size and distribution data
The sEV morphology was confirmed in a few samples by TEM (n = 4)
Fifty microliters of sEV suspension obtained after UC was fixed with 4% paraformaldehyde (PFA) (pH 7.2)
Five microliters of this suspension was placed on parafilm and a formvar/carbon grid
The grid was washed out with PBS and then fixed with 1% glutaraldehyde
embedded in methylcellulose-uranil and visualized with TEM (JEOL JEM 1010
Digital Micrograph software was used for image visualization and processing
This allows us to detect methylated DNA when the FAM probe initiates amplification
even if it occurs significantly later than the amplification of the VIC probe (representing unmethylated DNA) as expected
This approach enables the accurate quantification of tumor DNA methylation levels in blood samples with higher sensitivity and precision
a subgroup analysis of OS was performed among those patients who underwent incomplete resection
The statistical analysis was carried out using R (version 4.0.2) and survival
This study was conducted under the approval of the ethics committee of the La Paz University Hospital with the ethics number PI-2887
The study was performed according to the Helsinki Declaration
Informed consent was signed by all patients
</a>Diagram of patients included in the study and collected samples. The general characteristics of the patients are shown in Table <a data-track="click" data-track-label="link" data-track-action="table anchor" href="/articles/s41598-024-62061-8#Tab1">1</a>, while the methylation results derived from the paired samples are shown in Table <a data-track="click" data-track-label="link" data-track-action="table anchor" href="/articles/s41598-024-62061-8#Tab2">2</a>
NTA analysis was performed on those samples where there was sufficient plasma to perform both determinations, MGMT methylation and NTA assays. We also confirmed the presence of sEVs by EM. The use of both methods confirmed that most of the EVs observed were less than 200 nm in size and had the characteristic cup shape of sEVs (Fig. <a data-track="click" data-track-label="link" data-track-action="figure anchor" href="/articles/s41598-024-62061-8#Fig2">2</a>).
</a>Extracellular vesicle characterization. (A) Visualisation of circulating sEVs obtained from the plasma of GB patients using NTA. Examples of extracellular vesicle size distribution profiles in four out the fourteen samples obtained by NTA. NTA confirms the presence of extracellular vesicles below 150 nm. ID36 and ID37 are used to identify patients according to Table <a data-track="click" data-track-label="link" data-track-action="table anchor" href="/articles/s41598-024-62061-8#Tab2">2</a>
(B) Visualisation of circulating sEVs obtained from the plasma of GB patients using transmission electron microscopy (EM)
Images were taken at a magnification of 120,000
EM images showed small vesicles of approximately 120–130 nm in diameter
The arrow indicates a larger vesicle (&gt; 200 nm)
As expected, mMGMT methylation was not detected in the sEV DNA of any of the eight healthy donors (Supplementary Fig. <a data-track="click" data-track-label="link" data-track-action="supplementary material anchor" href="/articles/s41598-024-62061-8#MOESM1">1</a>)
six corresponded to patients who had undergone macroscopically complete resection as described intraoperatively by the neurosurgeon
Complete resection was confirmed in two of these patients (ID22 and ID44) by early postoperative magnetic resonance imaging (MRI)
Although the targeted surgical lesion was completely resected for patient ID21
where a known residual lesion was confirmed by postoperative MRI
The extent of resection could not be confirmed by early postoperative MRI for patient ID45
given that images were not assessable due to the patient’s movements when the images were taken
although intraoperative complete resection was described for patient ID46
tumor remnants were observed in the early postoperative MRI
Postsurgical MRI was not available for the final patient
The kappa index showed moderate concordance (k = 0.580) between tumor tissue-DNA and sEV-DNA (p &lt; 0.001)
If we only considered the subgroup of patients who underwent incomplete resection or biopsy
we achieved a concordance of 90.0% (95% CI: 68.3–98.8)
a sensitivity to detect methylation in sEV-DNA of 87.5% (95% CI: 47.3–99.7) and a specificity of 91.7% (95% CI: 61.5–99.8)
with a PPV of 87.5% (95% CI: 51.3–97.9) and an NPV of 91.7% (95% CI: 63.6–98.6)
The kappa index was 0.792 (substantial concordance) between tumor tissue-DNA in the subgroup of patients who underwent incomplete resection or biopsy and sEV-DNA (p &lt; 0.001)
</a>Overall Survival curves in our patient cohort
(A) Kaplan–Meier estimates of overall survival according to MGMT methylation in tumor sample
(B) Kaplan–Meier estimates of overall survival according to MGMT methylation in sEVs-DNA
(C) Kaplan–Meier estimates of overall survival according to MGMT methylation in sEVs-DNA in the subgroup of patients without complete resection
This result led us to evaluate the median OS in the group of patients who did not undergo complete resection, in which a poorer OS was achieved in patients with MGMT-unmethylated versus MGMT-methylated sEVs-DNA (10.9 months; 95% CI: 5.3–16.5 versus not reached (p value not assessable due to lack of events) (Fig. <a data-track="click" data-track-label="link" data-track-action="figure anchor" href="/articles/s41598-024-62061-8#Fig3">3</a>C)
To study the implications of measuring mMGMT status in blood along the course of the disease, we collected 20 baseline and follow-up samples from eight patients (Fig. <a data-track="click" data-track-label="link" data-track-action="figure anchor" href="/articles/s41598-024-62061-8#Fig4">4</a>A).
</a>Serial monitoring sEVs-DNA methylation status
(A) sEVs-DNA methylation status of all samples from the 8 monitored patients
(B) Changes in MGMT methylation ∆Ct levels detected in sEVs-DNA-based liquid biopsy in patients ID1
(C) Changes in MGMT methylation ∆Ct levels detected in sEVs-DNA-based liquid biopsy in patients ID2
Patient ID28 had a similar course to patient ID1
it was not possible to obtain further follow-up samples to confirm a favorable evolution
Patient ID32 had a similar situation to the previous patient
the dynamics of mMGMT in sEV-DNA were closely associated with the MRI results and patient prognoses
providing a valuable tool for the follow-up of patients with mMGMT
Patient ID34 underwent complete resection of a methylated MGMT tumor
The methylation remained at the baseline sEV-DNA sample
but was no longer observed in the remaining follow-up
MRI performed one month before sample collection showed no progression; however
the patient progressed and died two months after the sample was obtained; therefore
it was not possible to obtain a second follow-up at that stage to corroborate the mMGMT status
Patients ID12 and ID41 had an absence of mMGMT in the tumor and
who underwent complete resection of an unmethylated MGMT GB tumor
sEV-DNA methylation in the baseline sample was detected
the patient no longer showed mMGMT and remained stable until 9 months
The comprehensive characterization obtained using these techniques confirms the successful isolation of EVs
While we acknowledge the potential presence of platelets or lipoproteins
we maintain that this inherent limitation does not compromise the integrity of our results
which is exclusively associated with oncological processes
remains unaffected by these considerations
it is noteworthy to mention that no studies evaluating the methylation of MGMT using DNA in EVs have been identified to date
our study provides an unprecedented increase in the percentage of marker agreement between tissue and plasma
reaching 90% concordance with a sensitivity of 87.5% and a specificity of 92% in subtotal resection cases
given the primary objective of a predictive biomarker is forecast treatment response
our results suggest that MGMT methylation in tumor approaches statistical significance in predicting survival although our limited sample size
patients undergoing subtotal resection may also benefit from the mMGMT biomarker in EVs in terms of survival
albeit without reaching statistical significance
While acknowledging the limitations of the limited sample size
our data suggest that mMGMT in EVs may have potential significance as a predictive biomarker in glioblastoma
further investigation is needed to support our observations and facilitate the integration of the mMGMT biomarker into clinical contexts
The potential increase in sensitivity underscores the critical role of timing of sample collection and highlights the importance of optimizing sample collection protocols to increase the potential utility of liquid biopsy approaches
each of these studies employed varying sample types (serum/plasma)
it is paramount to exercise caution when interpreting and comparing absolute values
recognizing the inherent limitations associated with these methodological disparities
we recruited twenty samples from eight patients with baseline and at least one follow-up sample
We found that patients with mMGMT GB harboring detectable mMGMT at baseline and during follow-up in blood samples were patients who showed tumor progression
patients with mMGMT in sEV-DNA in the baseline sample that decreased in the follow-up samples were patients with radiographic evidence of tumor response
changes in ∆Ct levels in sEV-DNA provide relevant predictive information to manage and monitor the outcome of patients with GB
the in vitro study conducted by Maire et al
together with our in vivo data provide the basis for the diagnostic characterization of genome-wide methylation in sEV-DNA
which would allow the molecular classification and disease monitoring of CNS tumors in patients in whom tumor samples are not available
We report for the first time the detection of mMGMT in sEV-DNA with a sensitivity and specificity of 87.5% and 90%
The presence of MGMT in sEV-DNA appeared to show an association with patient OS and facilitated patient monitoring; however
statistical power limitations prevent a definitive determination
the results obtained here represent an important contribution to the field of extracellular vesicle-based liquid biopsy
suggesting that it reflects the heterogeneity of tumor tissue and represents a promising tool for biomarker detection
further larger studies are required to confirm our findings
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request
The 2016 World Health Organization Classification of Tumors of the Central Nervous System: A summary
The 2021 WHO Classification of Tumors of the Central Nervous System: A summary
Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma
High-grade glioma: ESMO Clinical Practice Guidelines for diagnosis
Inactivation of the DNA-repair gene MGMT and the clinical response of gliomas to alkylating agents
Prognostic or predictive value of MGMT promoter methylation in gliomas depends on IDH1 mutation
Clinical validation of a novel quantitative assay for the detection of MGMT methylation in glioblastoma patients
Intratumoral heterogeneity identified at the epigenetic
genetic and transcriptional level in glioblastoma
Liquid biopsy: Monitoring cancer-genetics in the blood
Liquid biopsy in gliomas: A RANO review and proposals for clinical applications
Integrating liquid biopsies into the management of cancer
Detection of circulating tumor DNA in early- and late-stage human malignancies
Pyrosequencing versus methylation-specific PCR for assessment of MGMT methylation in tumor and blood samples of glioblastoma patients
MGMT promoter methylation in serum and cerebrospinal fluid as a tumor-specific biomarker of glioma
Alu hypomethylation and MGMT hypermethylation in serum as biomarkers of glioma
DNA sequences within glioma-derived extracellular vesicles can cross the intact blood-brain barrier and be detected in peripheral blood of patients
Performance status assessment by using ECOG (Eastern Cooperative Oncology Group) score for cancer patients by oncology healthcare professionals
Modified criteria for radiographic response assessment in glioblastoma clinical trials
Minimal information for studies of extracellular vesicles 2018 (MISEV2018): A position statement of the International Society for Extracellular Vesicles and update of the MISEV2014 guidelines
Impact of biofluid viscosity on size and sedimentation efficiency of the isolated microvesicles
Tumor cell-specific BRCA1 and RASSF1A hypermethylation in serum
and peritoneal fluid from ovarian cancer patients
MAFG is a potential therapeutic target to restore chemosensitivity in cisplatin-resistant cancer cells by increasing reactive oxygen species
Reporting Recommendations for Tumor Marker Prognostic Studies (REMARK): Explanation and elaboration
ESMO recommendations on the use of circulating tumour DNA assays for patients with cancer: A report from the ESMO Precision Medicine Working Group
The measurement of observer agreement for categorical data
Interrater reliability: The kappa statistic
Extent of resection and survival in patients with glioblastoma multiforme: Systematic review and meta-analysis
Association of maximal extent of resection of contrast-enhanced and non-contrast-enhanced tumor with survival within molecular subgroups of patients with newly diagnosed glioblastoma
Association of the extent of resection with survival in glioblastoma: A systematic review and meta-analysis
Effect of extent of resection on survival of patients with glioblastoma
WHO Grade 4 (WHO 2021): Systematic review and meta-analysis
Definition of the prognostic role of MGMT promoter methylation value by pyrosequencing in newly diagnosed IDH wild-type glioblastoma patients treated with radiochemotherapy: A large multicenter study
Clinical utility of comprehensive cell-free DNA analysis to identify genomic biomarkers in patients with newly diagnosed metastatic non-small cell lung cancer
Cerebrospinal fluid tumor DNA for liquid biopsy in glioma patients’ management: Close to the clinic?
The potential of cerebrospinal fluid-based liquid biopsy approaches in CNS tumors
Blood–brain barrier disruption: A common driver of central nervous system diseases
MIBlood-EV: Minimal information to enhance the quality and reproducibility of blood extracellular vesicle research
and p14ARF promoter methylation in circulating tumor-derived DNA of central nervous system cancer patients
Serum DNA can define tumor-specific genetic and epigenetic markers in gliomas of various grades
Quantitative detection of multiple gene promoter hypermethylation in tumor tissue
and cerebrospinal fluid predicts prognosis of malignant gliomas
Methylated tumor-specific DNA as a plasma biomarker in patients with glioma
O6-methyl-guanine-DNA methyltransferase methylation in serum and tumor DNA predicts response to 1,3-bis(2-chloroethyl)-1-nitrosourea but not to temozolamide plus cisplatin in glioblastoma multiforme
Extracellular-vesicle-based cancer panels diagnose glioblastomas with high sensitivity and specificity
High prevalence of mutant KRAS in circulating exosome-derived DNA from early-stage pancreatic cancer patients
Circulating nucleic acids are associated with outcomes of patients with pancreatic cancer
Exosomes as a biomarker platform for detecting epidermal growth factor receptor-positive high-grade gliomas
Exosomes as novel diagnostic biomarkers and therapeutic tools in gliomas
Longitudinal bottom-up proteomics of serum
and cerebrospinal fluid reveals candidate biomarkers for early detection of glioblastoma in a murine model
Intratumor heterogeneity and branched evolution revealed by multiregion sequencing
Clinical utility of plasma cell-free DNA in adult patients with newly diagnosed glioblastoma: A pilot prospective study
Cell-free DNA and circulating TERT promoter mutation for disease monitoring in newly-diagnosed glioblastoma
TERT promoter mutation analysis for blood-based diagnosis and monitoring of gliomas
Prognostic significance of gene expression and DNA methylation markers in circulating tumor cells and paired plasma derived exosomes in metastatic castration resistant prostate cancer
Genome-wide methylation profiling of glioblastoma cell-derived extracellular vesicle DNA allows tumor classification
<a data-track="click" data-track-action="download citation references" data-track-label="link" rel="nofollow" href="https://citation-needed.springer.com/v2/references/10.1038/s41598-024-62061-8?format=refman&amp;flavour=references">Download references</a>
The authors thank Morote translation services for their assistance with language editing and the HULP-IdiAPZ Biobank
The information provided in this study is included in a patent application process (EP19382299.6); therefore
based on the purposes of this paper and should not be published if it does not respond to the purpose thereof
This study was funded by Instituto de Salud Carlos III (ISCIII) and co-funded by the European Union
Grants numbers: PI18/000050; PI21/000145; DTS20/000029; JR17/000016; JR21/000003; FI19/000061; CIBERONC-CB16/12/0295
GHETTI2021 and Instituto de Salud Carlos III (ISCIII)
del Programa FORTALECE del Ministerio de Ciencia e Innovación
Rocío Moreno-Velasco &amp; Inmaculada Ibánez de Cáceres
Biomarkers and Experimental Therapeutics in Cancer
Javier de Castro &amp; Inmaculada Ibánez de Cáceres
Translational Medical Oncology Group (ONCOMET)
University Clinical Hospital of Santiago (CHUS/SERGAS)
Angel Díaz-Lagares &amp; Rafael López-López
Centro de Investigación Biomédica en Red de Cáncer (CIBERONC)
University Hospital Complex of Santiago de Compostela
Catalina Vivancos Sánchez &amp; Maria Luisa Gandía-González
Instituto de Investigaciones Biomédicas ‘Alberto Sols’ (CSIC-UAM)
Methodology: RRA (sample collection and processing)
VM (patient recruitment and clinical data collection)
CVS (patient recruitment and clinical data collection)
MLGG (patient recruitment and clinical data collection)
JDC (patient recruitment and clinical data collection)
The authors declare no competing interests
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations
<a data-test="citation-link" data-track="click" data-track-action="download article citation" data-track-label="link" data-track-external="" rel="nofollow" href="https://citation-needed.springer.com/v2/references/10.1038/s41598-024-62061-8?format=refman&amp;flavour=citation">Download citation</a>
DOI: https://doi.org/10.1038/s41598-024-62061-8
Anyone you share the following link with will be able to read this content:
a shareable link is not currently available for this article
Sign up for the Nature Briefing: Cancer newsletter — what matters in cancer research
Juan Romo&nbsp;is Rector of Universidad Carlos III de Madrid (UC3M)
The University serves as the United Nations Academic Impact hub for Sustainable Development Goal 11 (Sustainable cities and communities)
issued by United Nations Secretary-General António Guterres
emphasizes that vulnerable sectors of urban societies are especially harmed both by the incidence of the virus and the economic impact of related shutdown measures
These figures reflect structural problems that are difficult to tackle in the short term
residents of relatively poor areas are mainly employed in sectors that offer limited opportunities for distanced work
Their mobility depends mainly on public transportation
which tends to become overcrowded during rush hours
These residents experience long commutes from peripheral locations to their places of work
thus increasing their risk of contracting the virus
They live in smaller and relatively overcrowded dwellings
which makes self-isolation more difficult and contagion more likely
They are also greatly affected by the digital divide
which limits their access to remote services that facilitate working from home
As Spain heads inexorably towards a second wave of the pandemic
with the virus spreading among younger age groups
epidemiological data show that Madrid’s poorest districts are disproportionately affected
What can be done? As <a href="https://www.un.org/press/en/2020/sgsm20189.doc.htm" rel="noreferrer noopener" target="_blank">stressed</a> by the Secretary-General
we should not miss the opportunity that COVID-19 has afforded us to rethink how we live and interact in our cities
To build back better urban environments we need to move towards new models of urban mobility that reduce the risk of contagion and make cities more resilient
sustainable and inclusive at the same time
Public policies should promote working from home whenever possible to reduce unnecessary travel to workplaces
car sharing and alternative transportation options
careful urban planning can gradually transform this goal into a reality
we promote an interdisciplinary approach that combines technical
Sustainable mobility is also an essential part of our management
With four campuses spread over an extended territory
our University conducts regular surveys of mobility patterns of its students and employees
subsidizes the use of public transportation and offers parking stations for bikes and e-bikes
as well as charging stations for electric vehicles
non-governmental organizations and private firms in the identification of problems related to urban mobility and environmental sustainability
Breaking down barriers between universities and society and promoting an active engagement with cities also inspire our participation in <a href="https://yufe.eu/" rel="noreferrer noopener" target="_blank">Young Universities for the Future of Europe</a> (YUFE)
This is an alliance of universities in 11 countries
whose principles include the development of university-citizen communities and the promotion of multidisciplinary
intersectoral approaches to addressing societal challenges
YUFE partners are committed to building strong quadruple-helix models of open innovation
and want to bring the challenge one step further for the mutual benefit of all European citizens
The more universities succeed in their ambitions
the more our societies will be able to produce innovative and sustainable solutions to the challenging problems of our cities in the post-COVID-19 world
The UN Chronicle is not an official record
It is privileged to host senior United Nations officials as well as distinguished contributors from outside the United Nations system whose views are not necessarily those of the United Nations
in maps or articles do not necessarily imply endorsement or acceptance by the United Nations
more efficient systems and cleaner alternatives in agriculture can reduce greenhouse gas&nbsp;emissions and air pollution
As we mark International Girls in ICT Day,&nbsp;let’s make sure that every young woman and girl has the chance to be meaningfully connected with ample opportunities to become digital transformation leaders
which is why they have been an important part of the communications strategy at the United Nations
Welcome to www.telcotitans.com. This site uses cookies. <a href="https://www.telcotitans.com/privacy-and-cookie-policy/1788.article">Read our policy</a>
<a href="https://link.telcotitans.net/blueplanet"></a>
<a href="https://link.telcotitans.net/celfocus"></a>
<a href="https://link.telcotitans.net/snowflake"></a>
<a href="https://link.telcotitans.net/blue-planet"></a>
 <a href="https://www.telcotitans.com/samritee-kumari/1026.bio" rel="author">Samritee Kumari</a>2022-03-03T09:52:00
Already registered?&nbsp;<a href="https://www.telcotitans.com/sign-in">Sign in here</a>
By&nbsp;<a class="ad-append-return" href="https://account.telcotitans.com/register">registering</a>&nbsp;for a free account,&nbsp;you will get immediate&nbsp;access to the rest of this article
<a class="ad-append-return" href="https://account.telcotitans.com/register">Register now</a>
&nbsp;Want full access to TelcoTitans content? Check out our&nbsp;<a href="https://www.telcotitans.com/subscriptions">subscription options</a>
Site powered by <a href="https://www.abacusemedia.com/webvisioncloud">Webvision Cloud</a>
<a href="https://videoweek.com/wp-content/uploads/2023/03/Mediaset-Spain.jpg"></a>
Mediaset España shareholders have approved the company’s takeover by MediaForEurope (MFE)
The Spanish commercial broadcaster is expected to be absorbed into MFE within one month
shareholders agreed to place the business in a new company as an intermediary measure
Grupo Audiovisual Mediaset España (GA Mediaset) will be headed by Borja Prado as president and Alessandro Salem and Massimo Musolino as co-CEOs
&#8220;We believe that the merger [makes] perfect sense from a strategic point of view
since it constitutes not only a logical evolution but also the safest path for the survival and expansion of the company,&#8221; said Prado
with dissenting shareholder Jose Antonio del Barrio Colmenarejo accusing the Board of Directors of collecting future compensation from the deal
which undervalues the company and deprives minority shareholders
The debate over MFE&#8217;s alleged infringement of shareholder rights has been raging since the Berlusconi-owned broadcasting group was first proposed in 2019
leading to a series of court battles between MFE and French media giant Vivendi
Buying out the remaining stake in Mediaset España is a core part of the holding group&#8217;s plan to become a pan-European media powerhouse
It has faced resistance from the Spanish outfit and failed to secure the 90 percent of shares required to delist Mediaset España from the Madrid stock exchange
But last year MFE gained 83 percent control of the broadcaster
and secured a waiver on the minimum acceptance threshold
This week MFE acquired Vivendi&#8217;s 1.05 percent stake in the Spanish outfit
effectively allowing it to delist the company&#8217;s shares
MFE has been raising its stake in ProSiebenSat.1 to become the German broadcaster&#8217;s largest shareholder
MFE is ultimately expected to launch a takeover bid for ProSieben
though Berlusconi has ruled this out for the time being
Mediaset España&#8217;s shares have fallen by around 67 percent over the course of the drama
prompting accusations that the Board of Directors deliberately lowered the valuation in light of possible offers from MFE
as well as the alleged ousting of Paolo Vasile after 23 years as CEO
Company president Borja Prado denies the accusations
pointing to &#8220;market circumstances&#8221; behind the company&#8217;s plummeting share price
He assured shareholders that the merger will create a larger
diversified media group with access to a combined audience of over 100 millions viewers
&#8220;Both Deutsche Bank and Banco Santander
issued their confirmed report that the [evaluation] is adequate from a financial point of view,&#8221; he added
Mediaset España saw its revenues drop 1.3 percent last year
citing the contraction of Spain&#8217;s linear TV ad market
Meanwhile MFE-controlled Mediaset Italia witnessed a 1.1 percent increase in revenues
though those numbers only account for the first half of the year
Follow VideoWeek on <a style="color: color 4;" href="https://twitter.com/videoweek">Twitter</a> and <a style="color: color 4;" href="https://www.linkedin.com/company/videoweek/">LinkedIn</a>
Join thousands of video and CTV advertising professionals by signing up to our newsletter
<a href="https://videoweek.com/privacy-policy/">Privacy Policy</a>
<a href="/articles/s41598-021-81205-8/metrics" data-track="click" data-track-action="view metrics" data-track-label="link" rel="nofollow">Metrics details</a>
The increased prevalence of childhood obesity is expected to translate in the near future into a concomitant soaring of multiple cardio-metabolic diseases
that includes multiple and multidomain potential risk factors: genetics
all these factors are expected to exert their influence through a specific and especially convoluted way during childhood
Machine Learning methods are the appropriate tools to model this complexity
given their ability to cope with high-dimensional
we have analyzed by Machine Learning a sample of 221 children (6–9 years) from Madrid
Both Random Forest and Gradient Boosting Machine models have been derived to predict the body mass index from a wide set of 190 multidomain variables (including age
A consensus relative importance of the predictors has been estimated through variable importance measures
implemented robustly through an iterative process that included permutation and multiple imputation
We expect this analysis will help to shed light on the most important variables associated to childhood obesity
in order to choose better treatments for its prevention
This is especially interesting for explanatory purposes of the predicted endpoint
Another robust ensemble-based ML method is Gradient Boosting Machines (GBM)<a data-track="click" data-track-action="reference anchor" data-track-label="link" data-test="citation-ref" aria-label="Reference 11" title="Friedman, J. Greedy function approximation: A gradient boosting machine. Ann. Stat. 29, 1189–1232 (2001)." href="/articles/s41598-021-81205-8#ref-CR11" id="ref-link-section-d487460053e489">11</a>
where one tree is fitted to reduce the prediction error of the previous ones
Normally a stochastic version of this approach is used
using at each new added tree a random subsample (e.g
in order to decorrelate the trees and thus result in predictions with less variance
GBM are also amenable to perform variable importance calculations
and no variable importance techniques were used to rank the predictors
Children obesity has peculiarities that make it to require specific modeling efforts
due to the huge hormonal and metabolic changes that occur in this period
there is a lack of ML models for pediatric samples and with high-dimensional
especially those focused on estimating the relative importance of these variables
although in that case the set of variables is more restricted both in number and domains
mostly of psychological nature and with no genetic data
for this work we set out to analyze a pediatric sample by ML and predict its BMI based on a large set of 190 variables from different domains: single nucleotide polymorphisms (SNPs)
The sample was a group of schoolchildren of Madrid (Spain) enrolled in the GENYAL study for the prevention of childhood obesity
and here we perform a cross-sectional analysis of the baseline data
we attempted to rank the variables and identify those more strongly associated with the target
in order to better characterize the important features for children obesity
in order to assess the robustness of the estimated ranks
and derived a consensus variable importance score for all the predictors by combining the predictions of the two models
This consensus ranking will assist in developing better prevention strategies that will result in better expectations for quality of life and longevity in the future
We have to stress at this point that we use here the term “predictor variable” in an statistical sense
where the values of one or more independent or predictor variables are used to obtain the value (predict) for a dependent variable (in this case BMI)
Given the cross-sectional nature of the data
we are actually modelling associations of BMI with other variables at a given point in time
and not forecasting future values of BMI given some current values of the independent variables
Table <a data-track="click" data-track-label="link" data-track-action="supplementary material anchor" href="/articles/s41598-021-81205-8#MOESM1">S1</a> (Supplementary Material) collects the 190 predictor variables used in the analysis
They are grouped in different domains: characteristics of schoolchildren (3); genetics (1
from 11 SNPs); physical and leisure activities (24); diet
food and nutrients (80); risk factors of pregnancy and birth (39); social
The average age of the 221 participants was 6.75 ± 0.73 years (52.50% were girls (n = 116) and 47.50% boys (n = 105))
32.2% of the schoolchildren evaluated had excess weight (EW) (18.1% overweight and 14.1% obesity)
These figures were 25.4% and 19.0% when the International Obesity Task Force (IOFT) standard or the national criteria of the Orbegozo Foundation were used
Table <a data-track="click" data-track-label="link" data-track-action="table anchor" href="/articles/s41598-021-81205-8#Tab1">1</a> shows the main descriptive characteristics regarding the schoolchildren families
Regarding the nutritional status of the parents
57.5% of the fathers and 30.4% of the mothers had EW
The main diet, physical activity and birth characteristics of schoolchildren by sex are summarized in Table <a data-track="click" data-track-label="link" data-track-action="table anchor" href="/articles/s41598-021-81205-8#Tab2">2</a>
The variants distribution of the set of SNPs selected for the genetic risk score (GRS, see <a data-track="click" data-track-label="link" data-track-action="section anchor" href="/articles/s41598-021-81205-8#Sec8">Methods</a>) are presented in Table <a data-track="click" data-track-label="link" data-track-action="table anchor" href="/articles/s41598-021-81205-8#Tab3">3</a>
These gene variants were consistent with the Hardy–Weinberg equilibrium in all the cases (p-values ≥ 0.05)
</a>Scatter plot of the average predicted vs observed BMI for the RF models
</a>Variable importance plot of the top-30 predictor variables for the GENYAL sample, according to the RF models to predict childhood BMI.
</a>Mean average rank and 95% confidence intervals of the 30 most important predictor variables from the RF models to predict childhood BMI
The five most important variables are (in this order): Familiar nutri-status perception (Perception of the person completing the questionnaire about child's nutritional status) &gt; Relation TEI-TEE (%) (Percentage of difference between Total Energy Intake (TEI) and Total Energy Expenditure (TEE)) &gt; BMI of the father &gt; BMI of the mother &gt; Mother’s Meals (number of daily food servings of the mother)
with both Familiar nutri-status perception and Relation TEI-TEE (%) having a null confidence interval in their average rank
as in all the imputations they were the first- and second-most important variables
The BMI of both parents share the same narrow confidence interval (3–4)
while Mother’s Meals had a slightly larger confidence interval (5–7)
The next-important variables (in decreasing importance) are IPAC (Individual Physical Activity Coefficient) &gt; GRS (genetic risk score) &gt; Vit D (Vitamin D (mcg): quantity of daily vitamin D intake) &gt; Mother's disease: HTG (Mother has hypertriglyceridemia by medical diagnose)
with increasingly larger confidence intervals: (5–7)
The following variables show much larger confidence intervals
so that although on average they show an increasing rank
their ranking for new samples is expected to be less well defined
</a>Scaled relative importance plot of the top-30 predictor variables for the GENYAL sample
</a>Consensus Variable Importance plot of the top-30 predictor variables for the GENYAL sample
after the RF &amp; GBM models to predict childhood BMI
in view of the crucial role that prevention plays to control the high obesity prevalence
the identification of its most important risk factors could help to develop effective nutritional and educational intervention strategies
we attempted to rank a wide set of 190 predictor variables from different domains in order to predict the BMI of children by means of ML models of the RF and GBM types
there is no parallel in the literature in this regard by this use of such a large multidomain set of variables for childhood obesity
our questionnaire-based TEI and TEE do contain valuable information about the energy input and expenditure
and thus the Relation TEI-TEE (%) variable results in one of the best predictors for BMI
namely BMI percentile change after 10 years in adolescent girls
as well as transition from normal weight to overweight or obesity
The predictor variable set was more limited (41 variables) and with a more restricted set of domains: diet
appeared within the most important variables; this is probably not unexpected
given that the sample was composed of adolescent girls
were this domain would be of more importance
We think that this domain would be of less importance in our 6–8 years old children
in pediatric samples it is frequent the use of age- and sex-specific BMI z-scores instead of raw BMI
our sample has a very narrow distribution of ages
with 84% of the children being 6–7 years old
and we did not observe significant differences between the two sexes
as the z-scores are quite dependent on the population they are based on
this information should be interpreted with caution
but it is important to consider that this study is framed in an intervention study of five years and corresponds to a baseline cross-sectional analysis
at this point this model was derived for explanatory purposes
in order to identify the predictors most associated to BMI
The cross-sectional nature of the present baseline dataset prevents its use from demonstration of causality
This model rather suggests variables that would be important for childhood obesity
in order to be further tested in longitudinal settings
The new accumulated data along the study will be incorporated in order to derive models for predictive purposes to target appropriate preventive interventions to ameliorate effectively children obesity
our use of a permutation approach avoids overestimation of categorical variables with many classes
we removed highly-correlated variables that could also be overestimated
the picture obtained from the GBM analysis is rather similar to the RF one
with up to 20 the 30 top variables shared variables between the two methods
This gives confidence in the general conclusions above described about the influence of the different predictor variables
We must also take into account that many of these variables are correlated
so that the way that one method achieves a best fit will be different that the other given their different algorithms
while modeling basically the same physical mechanism
the important variable IPAC in the RF plot
while in the latter Active transport to school instead appears
a large component of the physical activity of the child (measured by IPAC) would be going to school walking or biking
and this is measured by the Active transport to school variable
In the GBM plot sleep time is the fifth most important predictor
there is a large similarity between the two descriptions of childhood obesity
taking into consideration that the dataset contains up to 190 predictor variables
it is worth highlighting the homogeneity of the sample in terms of distribution by sex and the absence of genetic relatedness and stratification (since the Hardy–Weinberg equilibrium is met by all the SNPs)
the sample shows a large representativeness with six schools from three different areas of the Community of Madrid involved
which allows to have a better knowledge of the situation throughout the Community and not from a specific school or area
GENYAL is a long-term clinical trial (ClinicalTrials.gov NCT03419520) for childhood obesity prevention
The duration of the project is planned to last 5 years (2017–2021) with annual data collection
including anthropometric and nutrigenetic assessment and questionnaires about physical activity
the main objective of GENYAL study was to design and validate a predictive model that identifies those children who would benefit most from actions aimed at reducing the risk of obesity and its complications through ML algorithms
The results shown in this paper corresponding to a cross section from data collected in the first year of the study (2017)
The research was approved by the Research Ethics Committee of the IMDEA Food Foundation (PI:IM024)
The study protocol follows the guidelines laid down in the Declaration of Helsinki and was performed in accordance with relevant regulations
All families signed their written informed consent to participate
Height was determined using a Leicester height rod with a millimetric accuracy (Biological Medical Technology SL
fat mass percentage and muscle mass percentage were assessed using a Body Composition Monitor (BF511- OMRON HEALTHCARE Co.
Waist circumference were taken using a non-elastic tape (KaWe Kirchner &amp; Wilhelm GmbH
an automatic digital monitor was used (OMRON M3-Intellisense) using a cuff suitable for children
The results of overweight and obesity rates were unified as a single category called excess weight (EW)
Parents’ BMI was calculated from the weight and height data reported by themselves
and genotype frequencies were in Hardy-Weingberg equilibrium (p &gt; 0.05)
The data obtained were processed and cleaned. Finally, a total of 190 variables obtained were classified into categories according to their specific nature. (Table <a data-track="click" data-track-label="link" data-track-action="supplementary material anchor" href="/articles/s41598-021-81205-8#MOESM1">S1</a>
These variables are described in what follows
Three variables were taken into account in this category: age
obtained from 11 SNPs variables well described as significant in childhood obesity
The GRS for each child was obtained as the sum of the number of risk alleles of each of the 11 SNPs over all the SNPs
by considering that each SNP can contain 0
the corresponding number of risk alleles would be 0
and analyzed using the DIAL software (Alce Ingeniería
Spain) in order to obtain information about macro and micronutrients
a questionnaire based on the surveys used in previous national studies (ALADINO and ELOIN) was used after receiving content validation by a group of Nutritionist
39 variables regarding the maternal and neonatal health and habits were obtained from self-reported ad hoc questionnaire completed by parents
This questionnaire was used after receiving content validation by a group of dietitians
43 variables were obtained from self-reported ad hoc questionnaire about the family’s status
R 3.4.2 (<a href="https://www.r-project.org/">https://www.r-project.org/</a>) was used for all the modeling and data analysis
The sample was initially characterized by a descriptive exploratory analysis
Qualitative data were presented as percentages and absolute frequencies while quantitative data were expressed as mean ± standard deviation
was applied in order to include multiple imputation in the variable importance estimation by taking into account both the between- and within-imputation variance in the importance scores
m = 1,…,M we estimated the average importance score of variable xj
where j = 1,…,p) as the average increase in the OOB MSE (Mean Squared Error) after OOB-permuting xj for each of the B trees of the RF a total of K times:
as well as the corresponding standard errors \({s}_{j}^{m}\)
From here the average importance score across the M imputations for each variable xj was obtained from:
the standardized importance score for each variable xj was calculated using:
where Vj is the weighted sum of the within (\({\stackrel{-}{W}}_{j}\)) and between (\({\stackrel{-}{B}}_{j}\)) imputation variances for variable xj:
The multiple imputation was also used to derive (rounded to the nearest integer) mean and 95% confidence intervals for the ranks of the importance scores of the different predictor variables in the RF models
where the summation is over the nonterminal nodes t of the J-terminal node tree T
\({v}_{j}\) is the variable selected for splitting in that node
1() is an indicator function that equals 1 if \({v}_{t}=j\) and 0 otherwise
and \({\widehat{i}}_{t}^{2}\) is the decrease of squared error associated to that variable
were successive base learners (regression trees in our case) are fitted to minimize the residuals of the previous one; therefore
the final relative importance’s for the GBM are obtained by averaging for each variable the relative importance’s over all the trees in the model
In order to derive a consensus variable importance’s
the two 100 imputations × 190 variable matrices of RF variable importance’s and GBM relative importance’s
were first min–max normalized (within each model) in order to make them comparable
the minimum and maximum average variable importance (relative importance for GBM) were used
the two matrices were merged and averaged for each predictor variable
The top-30 scoring variables were then plotted
Nutrition—EU Science Hub—European Commission. EU Science Hub <a href="https://ec.europa.eu/jrc/en/research-topic/nutrition" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="https://ec.europa.eu/jrc/en/research-topic/nutrition">https://ec.europa.eu/jrc/en/research-topic/nutrition</a> (2014)
Evaluating the evidence that the prevalence of childhood overweight is plateauing
Estudio ALADINO 2015: Estudio de Vigilancia del Crecimiento
Desarrollo Infantil y Obesidad en España 2015
Is obesity a problem among school children?
and prevention: An endocrine society clinical practice guideline
The epidemiology of obesity: A big picture
The elements of statistical learning: Data mining
Greedy function approximation: A gradient boosting machine
Machine learning techniques for prediction of early childhood obesity
A survey on utilization of data mining for childhood obesity prediction
in 8th Asia-Pacific Symposium on Information and Telecommunication Technologies 1–6 (2010)
Novak, B. &amp; Bigec, M. Application of artificial neural networks for childhood obesity prediction. in Proceedings 1995 Second New Zealand International Two-Stream Conference on Artificial Neural Networks and Expert Systems 377–380 (1995). doi:<a href="https://doi.org/10.1109/ANNES.1995.499512" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1109/ANNES.1995.499512">https://doi.org/10.1109/ANNES.1995.499512</a>
Novak, B. &amp; Bigec, M. Childhood obesity prediction with artificial neural networks. in Proceedings Ninth IEEE Symposium on Computer-Based Medical Systems 77–82 (1996). doi:<a href="https://doi.org/10.1109/CBMS.1996.507129" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1109/CBMS.1996.507129">https://doi.org/10.1109/CBMS.1996.507129</a>
Parameter identification and selection for childhood obesity prediction using data mining
A hybrid approach using Naïve Bayes and Genetic Algorithm for childhood obesity prediction
in 2012 International Conference on Computer Information Science (ICCIS) vol
Comparing data mining methods with logistic regression in childhood obesity prediction
Predicting childhood obesity using electronic health records and publicly available data
Developing an algorithm to detect early childhood obesity in two tertiary pediatric medical centers
The relative importance of predictors of body mass index change
overweight and obesity in adolescent girls
Machine learning models to predict childhood and adolescent obesity: A review
Munger, E. et al. Application of machine learning to determine top predictors of non-calcified coronary burden in psoriasis: An observational cohort study. J. Am. Acad. Dermatol. <a href="https://doi.org/10.1016/j.jaad.2019.10.060" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1016/j.jaad.2019.10.060">https://doi.org/10.1016/j.jaad.2019.10.060</a> (2019)
Predicting attention-deficit/hyperactivity disorder severity from psychosocial stress and stress-response genes: A random forest regression approach
A novel surgical predictive model for Chinese Crohn’s disease patients
Artificial intelligence and machine learning in endocrinology and metabolism: The dawn of a new era
Blanchet, R., Kengneson, C.-C., Bodnaruc, A. M., Gunter, A. &amp; Giroux, I. Factors influencing parents’ and children’s misperception of children’s weight status: A systematic review of current research. Curr. Obes. Rep. <a href="https://doi.org/10.1007/s13679-019-00361-1" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1007/s13679-019-00361-1">https://doi.org/10.1007/s13679-019-00361-1</a> (2019)
Prevention of obesity and metabolic syndrome in children
Energy intake derived from an energy balance equation
and dual X-ray absorptiometry can provide acceptable caloric intake data among young adults
Estimation of energy expenditure using prediction equations in overweight and obese adults: a systematic review
Systematic review of the measurement properties of self-report physical activity questionnaires in healthy adult populations
On the origin of obesity: Identifying the biological
environmental and cultural drivers of genetic risk among human populations
A systematic examination of the association between parental and child obesity across countries
Viljakainen, H. et al. Genetic risk score predicts risk for overweight and obesity in Finnish preadolescents. Clin. Obes. 2, e12342. <a href="https://doi.org/10.1111/cob.12342" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1111/cob.12342">https://doi.org/10.1111/cob.12342</a> (2019)
Genetic risk clustering increases children’s body weight at 2 years of age—the STEPS Study
A new explained-variance based genetic risk score for predictive modeling of disease risk
Lambert, S. A., Abraham, G. &amp; Inouye, M. Towards clinical utility of polygenic risk scores. Hum. Mol. Genet. <a href="https://doi.org/10.1093/hmg/ddz187" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1093/hmg/ddz187">https://doi.org/10.1093/hmg/ddz187</a> (2020)
anthropometric and body composition variables between adolescents from 10 to 13 years old and their parents
Childhood obesity is a high-risk factor for hypertriglyceridemia: A case-control study in Vietnam
Diet quality and physical activity in relation to childhood obesity
Do physical activity and screen time mediate the association between European fathers’ and their children’s weight status
Cross-sectional data from the Feel4Diabetes-study
Commentary: Which child obesity definitions predict health risk?
Accuracy of body mass index in diagnosing obesity in the adult general population
Agreement between parent and child report of physical activity
sedentary and dietary behaviours in 9–12-year-old children and associations with children’s weight status
Beware Default Random Forest Importances. <a href="http://explained.ai/decision-tree-viz/index.html" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="http://explained.ai/decision-tree-viz/index.html">http://explained.ai/decision-tree-viz/index.html</a>
Bias in random forest variable importance measures: Illustrations
WHO. Physical status: the use and interpretation of anthropometry. <a href="http://www.who.int/childgrowth/publications/physical_status/en/" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="http://www.who.int/childgrowth/publications/physical_status/en/">http://www.who.int/childgrowth/publications/physical_status/en/</a>
Estudio de Crecimiento de Bilbao (Curvas y tablas de crecimiento
Establishing a standard definition for child overweight and obesity worldwide: International survey
WHO. Growth reference data for 5–19 years. <a href="http://www.who.int/growthref/en/" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="http://www.who.int/growthref/en/">http://www.who.int/growthref/en/</a>
The Q223R polymorphism of the leptin receptor gene as a predictor of weight gain in childhood obesity and the identification of possible factors involved
Diseño del estudio ELOIN y prevalencia de sobrepeso y obesidad en la población infantil de 4 años de la Comunidad de Madrid
Mediterranean Diet Quality Index in children and adolescents
Manual de nutrición clínica en atención primaria
Medicine, I. of. Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids. (2002). doi:<a href="https://doi.org/10.17226/10490" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.17226/10490">https://doi.org/10.17226/10490</a>
General principles for the collection of national food consumption data in the view of a pan-European dietary survey
Multiple imputation and random forests (MIRF) for unobservable
Statistical Analysis with Missing Data (John Wiley &amp; Sons
<a data-track="click" data-track-action="download citation references" data-track-label="link" rel="nofollow" href="https://citation-needed.springer.com/v2/references/10.1038/s41598-021-81205-8?format=refman&amp;flavour=references">Download references</a>
This study received logistical support by Conserjería de Educación e Investigación de la Comunidad de Madrid
The authors would like to acknowledge the help of all the teachers and head teachers of Juan Zaragüeta
Concepción Arenal y Rosa Luxemburgo schools; and
the families and children volunteers who participated in the investigation
We also acknowledge Dr María Ikonomopoulos for reviewing the English style
This research did not receive any specific grant from funding agencies in the public
These authors contributed equally: Helena Marcos-Pasero and Gonzalo Colmenarejo
Elena Aguilar-Aguilar &amp; Viviana Loria-Kohen
Molecular Oncology and Nutritional Genomics of Cancer
Production and Development of Foods for Health
Department of Production and Characterization of Novel Foods
was the principal investigator and was responsible for the study design
proposed the use of Machine Learning variable importance techniques and designed the RF
as well as developed the consensus score of the variables; G.R.R
supervised the final compilation of the manuscript and provided scientific advice and consultation
<a data-test="citation-link" data-track="click" data-track-action="download article citation" data-track-label="link" data-track-external="" rel="nofollow" href="https://citation-needed.springer.com/v2/references/10.1038/s41598-021-81205-8?format=refman&amp;flavour=citation">Download citation</a>
DOI: https://doi.org/10.1038/s41598-021-81205-8
Sign up for the Nature Briefing newsletter — what matters in science
<a href="/articles/s41598-020-64833-4/metrics" data-track="click" data-track-action="view metrics" data-track-label="link" rel="nofollow">Metrics details</a>
Environmental and genetic factors are associated with pandemic obesity since childhood
the association of overweight-obesity with these factors
We aimed here to assess the probabilities of being overweighed-obese in a randomly recruited cohort of Spanish children and adolescents (n = 415
5−17 years-old) by estimating the odds ratios for different predictor variables
and their relative importance in the prediction
adherence to the Mediterranean diet (KIDMED)
UM-B and UM-0) as biomarkers of the gut microbiota
and 53 single-nucleotide polymorphisms (SNPs) from 43 genes mainly related to obesity and cardiometabolic diseases
A proportional-odds logistic ordinal regression
While every variable was not independently associated with overweight-obesity
the ordinal logistic model revealed that overweight-obesity prevalence was related to being a young boy with either UM-B or UM-0
low KIDMED score and high contribution of a consortium of 24 SNPs
rs708272-CETP and rs2241766-ADIPOQ the top-ten contributing SNPs
Additional research should confirm and complete this model by including dietary interventions and the individuals’ gut microbiota composition
it is more conceivable to expect the complex action of a consortium of SNPs potentially interacting with many other variables and associated with different conditions
we aimed to assess the probabilities of being overweighed or obese in a cohort of children and adolescents from the Southeast of Spain by estimating the odds ratios (ORs) for different predictor variables and their relative importance in the prediction of the response
we considered as predictor variables the urolithin metabotypes as biomarkers of the gut microbiota
and a consortium of 53 SNPs from 43 genes mainly related to obesity and cardiometabolic diseases
The trial was registered at clinicaltrials.gov (NCT03318042)
and the Spanish National Research Council’s Bioethics Committee (Madrid
Inclusion criteria were ages from 5 to 17 years old and good health status
Exclusion criteria were diagnosed pathology
chronic medication and antibiotic intake one month before participating
A total of 415 children and adolescents were randomly recruited
Children within the 5 to 12 years old group (n = 202) were recruited from the public primary school ‘CEIP Jara Carrillo’ (Alcantarilla
Spain) and adolescents aged from 13 to 17 (n = 213) from the public high school ‘IES Alcántara’ (Alcantarilla
Parents were fully informed and gave their written informed consent before the participation of all students
Data analysis was made by TaqMan Genotyper Software v1.3 (autocaller confidence level &gt;90%)
only 21 predictors could be used to get reliable estimates
data reduction was applied to the SNPs by applying Multiple Correspondence Analysis (MCA) and using only the first 15 MCA dimensions
Ethnic groups representing less than 1% each were merged into the ‘Other’ category
The approximate βs and ORs (and their 95% confidence intervals
CI) of the remaining variables were reported
The full model was validated through bootstrap to provide estimates of the performance of the model in new data in comparison with the training data
The significant contributions (coefficients of determination R2 with p values &lt; 0.05) of SNPs to the essential MCA dimension were plotted to deconvolute it
We used the proportional odds assumption in the model
‘KIDMED’ and ‘Urolithin metabotype’ showed some deviation from this assumption
alternative extended continuation ratio models with this assumption relaxed for these variables did not result in better models according to the AIC
the higher complexity of the model was not compensated by the increase in the fit
The inclusion of transformations of some predictors
including a restricted cubic spline for both ‘Age’ and ‘KIDMED’
a genetic score (computed as the sum of risk alleles) was also tested as a possible surrogate for the SNPs variables
but this did not result in a better replacement for the MCA dimensions
Other statistical analyses were carried out using the SPSS software
followed by Bonferroni-corrected t-test (for post-hoc analysis) or the Kruskal–Wallis followed by Dunn’s test were used for normally and non-normally distributed data
respectively (KIDMED score vs FTO genotype TT
Comparison of non-normally distributed quantitative variables between two clusters was approached using the Mann-Whitney U-test (FTO TT genotype vs BMI or waist
Comparison of categorical variables was assessed using the Pearson’s χ2 test
Spearman’s rank or Pearson correlations were applied to explore possible associations between variables (BMI vs hip/height
Plots of data were performed using Sigma Plot 13.0 (Systat Software
Table&nbsp;<a data-track="click" data-track-label="link" data-track-action="table anchor" href="/articles/s41598-020-64833-4#Tab1">1</a> shows characteristics of the cohort as a function of age and sex
the distribution of urolithin metabotypes (A
physical activity and the percentage of normoweight
The participants were mainly Caucasian-Europeans (93.5%)
with a small proportion of Arabs (2.9%) and Amerindians (2.2%)
and a marginal presence of Black-Africans (0.96%)
</a>Distribution (%) of (A) BMI, (B) overweight–obesity (OW–OB), (C,D) BMI z-scores, and (E) urolithin metabotypes in the cohort (n = 415) from 5 to 17 years. (F) Distribution of urolithin metabotypes (UM-A, UM-B and UM-0) in normoweight (NW) and OW + OB children. The shadow area in C and D designates overweight-obesity according to WHO (<a href="https://www.who.int/growthref/who2007_bmi_for_age/en/">https://www.who.int/growthref/who2007_bmi_for_age/en/</a>)
Regarding the SNPs analysed, after correcting for multiple tests, three SNPs did not satisfy the Hardy-Weinberg equilibrium (HWE) (rs1801253, rs5082, rs11868035) (Supplementary Table&nbsp;<a data-track="click" data-track-label="link" data-track-action="supplementary material anchor" href="/articles/s41598-020-64833-4#MOESM1">1</a>)
The rest of the SNPs were in equilibrium and were close to European frequencies
we can speculate that the reason could be the association established by the presence of certain consanguinity (several sibling groups in the cohort)
since we aimed to compare variables from different domains and not only in estimating the particular effect of a single SNP
it is not absolutely necessary to have HWE in our approach
to estimate odds ratios of an MCA dimension and rank the predictors
</a>(A) KIDMED scores in girls (•) and boys (▴) from 5 to 17 years
(B) KIDMED scores (box plots) as a function of the genotypes (TT
AT and AA) of the rs9939609 SNP in the FTO gene
Significant differences are shown after the Kruskal Wallis test
Bar charts show the proportion of normoweight (NW) and overweight-obesity (OW + OB) in the cohort depending on the rs9939609-FTO genotype of the individuals
many exemptions prevented the usefulness of the combination of KIDMED scores and rs9939609 genotypes as unique predictors of overweight-obesity in this cohort
although both the urolithin metabotypes and rs9939609 SNP-FTO could partially contribute to the overweight-obesity distribution as independent predictor variables in this cohort of children and adolescents; however
all the possible SNP-SNP interactions together with the rest of variables had not been taken into account
we next developed an ordinal logistic model to identify the consortium of variables that could estimate the odds ratios of the overweight-obesity distribution in this cohort
we used as the ordinal response the normoweight
overweight and obesity classification for children
based on sex and age to estimate the odds ratios (and the corresponding 95% confidence intervals) for different predictor variables
as well as their relative importance in the prediction of the response
and genetic polymorphisms (44 SNPs were finally included and compressed into 15 MCA dimensions
overweight and obesity WHO-based categories was apparently adjusted by sex and age
we still observed a trend for decreasing average age when moving from normoweight
as well as enrichment in boys in the same order
we also included these variables in the model
</a>Apparent predictor importance in the ordinal logistic model
Variables are ranked by relevance (based on their χ2-df score)
The χ2 of the Wald test for each predictor is also shown
We next applied a fast-backwards approach based on AIC to obtain a reduced model and estimate the corresponding βs and ORs (as well as their 95% CI). The variables kept were ‘Sex’, ‘Age’, ‘Urolithin metabotype’, ‘KIDMED’, and the genetic components ‘SNP.Dim.3’, ‘SNP.Dim.11’, and ‘SNP.Dim.14’ (Table&nbsp;<a data-track="click" data-track-label="link" data-track-action="table anchor" href="/articles/s41598-020-64833-4#Tab2">2</a>)
The variable ‘Ethnicity’ was not retained in the simplified model
probably due to its high complexity (4 levels)
Our results reveal that being a boy either with UM-B or UM-0 and having a higher SNP.Dimension.14
all increased the chances of overweight-obesity in this study population
better adherence to the Mediterranean diet (higher KIDMED score)
and being UM-A was associated with lower probabilities of being overweighed-obese
all these variables operated additively to build the final probability of overweight-obesity for each subject
</a>Description of components of SNP Dimension-14
Significant R2 (p &lt; 0.05) for each SNP within the SNP Dimension-14 are shown
We used bootstrap to validate the model (Table&nbsp;<a data-track="click" data-track-label="link" data-track-action="table anchor" href="/articles/s41598-020-64833-4#Tab4">4</a>)
The model showed some degree of overfitting
reflected in a decrease of the indexes after correction for optimism
The discriminative capacity of the model was modest
although not negligible according to the Dxy
R2 indexes (it must be taken into account that the typical values of R2 in ordinal models are much lower than those observed in linear regression models)
The calibration showed some degree of shrinkage
as seen from the deviation of the intercept and slope from 0 and 1
the maximum calibration error in predicting p (Y &gt; normal)
we obtained a significant model that still showed predictive power in external data
there are no previous studies that explore the occurrence of overweight-obesity in children and adolescents by estimating the odds ratios for the predictor variables ‘urolithin metabotypes’ as gut microbiota biomarkers
and an identified consortium of 24 SNPs from 22 genes
mainly related to obesity and cardiometabolic diseases
that would be highly penalised by the AIC criterion
our sample was mainly of Caucasian-European origin (93.5%)
and it is expected that in a sample with a more balanced distribution of ethnicity
the modest predictive capability of the model suggested the need for additional predictors
given the multifaceted aetiology of obesity
allowed us to rank predictors of different domains by their importance
as well as to estimate odds ratios for them
and metabolic syndrome in a cohort of 577 Chinese subjects
all the above highlights again the need to consider SNPs consortia
instead of few SNPs in those studies aimed to associate SNPs with obesity
We are aware that the present study is an exploratory validation for a proof-of-concept
an ordinal logistic model that associates child overweight-obesity with a consortium of SNPs potentially interacting with the urolithin metabotypes-associated microbiota
Although we claim for the rationale of our approach and its potential usefulness
our results should be confirmed with additional research
We also acknowledge some limitations that should be considered in further studies
which also could improve its prediction capability
it would be interesting to include other possible variables
such as the detailed composition and functionality of the individuals’ gut microbiomes
and dietary interventions to evaluate not only associations but also individuals’ responses
The latter would be even better than the use of validated questionnaires
a higher number of SNPs (or many SNPs associated with a specific gene) should be explored
especially in children from other geographical origins and ethnicities
the inclusion of serobiochemical variables and traits related to obesity and its comorbidities (blood lipid profile
a parallel-group with all the children either normoweight or obese) should also be considered in further studies to confirm our model fully
The present research highlights the need for a holistic approach to unravel the predictors of overweight-obesity in children
in agreement with the multifaceted aetiology of obesity
the link of childhood overweight-obesity to multifactorial associations of environmental and genetic components
The ordinal logistic model revealed that child overweight-obesity prevalence was related to being a young boy with either UM-B or UM-0
it is of particular relevance the evaluation of interactive SNPs consortia along with the stratification of the children according to their urolithin metabotypes
of a dysbiotic-prone obesity-associated microbiota
All data generated or analysed during this study are included in this published article (and its Supplementary Information files)
Cardiovascular and metabolic consequences of obesity
Dissecting the role of the gut microbiota and diet on visceral fat mass accumulation
Media multitasking is associated with higher risk for obesity and increased responsiveness to rewarding food stimuli
A common variant in the FTO gene is associated with body mass index and predisposes to childhood and adult obesity
The genetics of obesity: FTO leads the way
The fat mass and obesity-associated (FTO) gene: obesity and beyond
A map of human genome sequence variation containing 1.42 million single nucleotide polymorphisms
Metabolic syndrome and altered gut microbiota in mice lacking Toll-like receptor 5
Can gut microbiota composition predict response to dietary treatments
The gut microbiota regulates white adipose tissue inflammation and obesity via a family of microRNAs
Being overweight or obese is associated with harboring a gut microbial community not capable of metabolizing the soy isoflavone daidzein to O-desmethylangolensin in peri- and post-menopausal women
The gut microbiota: a key factor in the therapeutic effects of (poly)phenols
the rescue of “old” metabolites to understand a “new” concept: metabotypes as a nexus among phenolic metabolism
The gut microbiota metabolism of pomegranate or walnut ellagitannins yields two urolithin-metabotypes that correlate with cardiometabolic risk biomarkers: comparison between normoweight
Obesity prevalence in relation to gut microbial environments capable of producing equol or O-desmethylangolensin from the isoflavone daidzein
Clustering according to urolithin metabotype explains the interindividual variability in the improvement of cardiovascular risk biomarkers in overweight-obese individuals consuming pomegranate: a randomized clinical trial
Acute benefits of the microbial-derived isoflavone metabolite equol on arterial stiffness in men prospectively recruited according to equol producer phenotype: a double-blind randomized controlled trial
The human gut microbial ecology associated with overweight and obesity determines ellagic acid metabolism
Polyphenols’ gut microbiota metabolites: bioactives or biomarkers
Deciphering the human gut microbiome of urolithin metabotypes: association with enterotypes and potential cardiometabolic health implications
The gut microbiota urolithin metabotypes revisited: the human metabolism of ellagic acid is mainly determined by aging
Targeted metabolic profiling of pomegranate polyphenols and urolithins in plasma
urine and colon tissues from colorectal cancer patients
Practical guide to measuring physical activity
Obesity genetics and cardiometabolic health: potential for risk prediction
Genome-wide association studies of obesity and metabolic syndrome
Genetic polymorphisms in adipokine genes and the risk of obesity: a systematic review and meta-analysis
Annotation of functional variation in personal genomes using RegulomeDB
HaploReg: a resource for exploring chromatin states
and regulatory motif alterations within sets of genetically linked variants
Association of calcium and dairy product consumption with childhood obesity and the presence of a Brain Derived Neurotropic Factor-Antisense (BDNF-AS) polymorphism
and survival analysis (2nd ed.) (Springer Series in Statistics
Efficient screening of non-normal regression models
A new look at the statistical model identification
Genetic predictors of weight loss in overweight and obese subjects
dietary intakes and anthropometric measures in European adults: The Food4Me study
Associations of the FTO rs9939609 and the MC4R rs17782313 polymorphisms with type 2 diabetes are modulated by diet
being higher when adherence to the Mediterranean diet pattern is low
The family environment and American adolescents’ risk of obesity as young adults
Overweight and obesity in a sample of children with autism spectrum disorder
obesity and girth of Australian preschoolers: prevalence and socio-economic correlates
Combined effect of FTO and MC4R gene polymorphisms on obesity in children and adolescents in Northwest China: a case-control study
and high blood pressure among adolescents: a cross-sectional study
The intestinal microbiota composition and weight development in children: the KOALA Birth Cohort Study
Pre-obese children’s dysbiotic gut microbiome and unhealthy diets may predict the development of obesity
Composition of gut microbiota and its association with body mass index and lifestyle factors in a cohort of 7-18 years old children from the American Gut Project
Gut microbiota and endothelial dysfunction markers in obese Mexican children and adolescents
Symptomatic atherosclerosis is associated with an altered gut metagenome
Ellagic acid metabolism by human gut microbiota: consistent observation of three urolithin phenotypes in intervention trials
Urolithin metabotypes can anticipate the different restoration of the gut microbiota and anthropometric profiles during the first year postpartum
An obesity-associated FTO gene variant and increased energy intake in children
and PPARG-2 polymorphisms with obesity traits and metabolic phenotypes in school-aged children
The functional significance of genetic variation within the beta-adrenoceptor
Association between obesity and a polymorphism in the α-1 adrenergic receptor gene (Gly389Arg ADRB1) in Caucasian women
β1-adrenoceptor gene polymorphism predicts long-term changes in body weight
Obesity is associated with the Arg389Gly ADRB1 but not with the Trp64Arg ADRB3 polymorphism in children from San Luis Potosí and León
Lipolysis and lipid mobilization in human adipose tissue
Adipocytes as a new source of catecholamine production
Urinary epinephrine and norepinephrine interrelations with obesity
and the metabolic syndrome in Hong Kong Chinese
<a data-track="click" data-track-action="download citation references" data-track-label="link" rel="nofollow" href="https://citation-needed.springer.com/v2/references/10.1038/s41598-020-64833-4?format=refman&amp;flavour=references">Download references</a>
This research was funded by the Project AGL2015-64124-R (MINECO
is the holder of a FPI predoctoral grant from MINECO (Spain)
Isabel Espinosa from the GENYAL platform (IMDEA-Food
Spain) for their help in the genotyping of the students
The authors are also grateful to Carlos Ortiz from the public primary school ‘CEIP Jara Carrillo’ (Alcantarilla
Teresa Saavedra and Ginés Bueno from the public high school ‘IES Alcántara’ (Alcantarilla
Spain) for being so well disposed towards this project and facilitating the recruitment process
We especially appreciate the participation of the students in this research and the willingness of their parents to make this possible
Núñez-Sánchez for their collaboration in the recruitment process
These authors contributed equally: Adrián Cortés-Martín and Gonzalo Colmenarejo
María Victoria Selma &amp; Juan Carlos Espín
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations
<a data-test="citation-link" data-track="click" data-track-action="download article citation" data-track-label="link" data-track-external="" rel="nofollow" href="https://citation-needed.springer.com/v2/references/10.1038/s41598-020-64833-4?format=refman&amp;flavour=citation">Download citation</a>
DOI: https://doi.org/10.1038/s41598-020-64833-4
<a href="/articles/s41598-018-26125-w/metrics" data-track="click" data-track-action="view metrics" data-track-label="link" rel="nofollow">Metrics details</a>
Plasmodium falciparum Standard Membrane Feeding Assay (PfSMFA) is the current gold standard mosquito based confirmatory transmission blocking (TrB) assay for human malaria
owing to its complexity only selected gametocytocidal molecules are progressed into SMFA
Predictive tools for evaluation of TrB behavior of compounds in SMFA would be extremely beneficial
but lack of substantially large data sets from many mosquito feeds preempts the ability to perform correlations between outcomes from in vitro assays and SMFA
a total of 44 different anti-malarial compounds were screened for inhibitory effect on male gamete formation in exflagellation inhibition assay (EIA) and the same drug-treated parasites were fed to mosquitoes in SMFA
Regression analysis was performed between outcomes of the two assays and regression models were applied to a randomly selected validation set of four compounds indicating no overfitting and good predictive power
the pIC50 for 11 different compounds obtained in the EIA was also correlated with pIC50’s in SMFA
Resulting regression models provided pIC50 predictions in SMFA with reasonably good accuracy thereby demonstrating the use of a simple in vitro assay to predict TrB of molecules in a complex mosquito based assay
The percentage exflagellation inhibition (EI) obtained from the EIA was calculated using the equation:
where EC and ET are the number of exflagellation centers per field in the control and in the compound-treated sample
The same treated and un-treated gametocyte cultures which were used in EIA were fed to mosquitoes in SMFA and the output was measured by enumeration of midgut oocysts at seven days post-feeding
by considering two different infection outcomes; (i) total number of P
falciparum oocysts per mosquito midgut (oocyst intensity) and its mean was estimated
and (ii) the total number of mosquitoes infected or prevalence of infection
The percentage reduction of mean oocyst intensity (OR) was defined as
The percentage reduction in the prevalence of infection or block in transmission (BIT)
Using EIA and SMFA data from these 44 different compounds
corresponding to a total of 148 sample points
we performed regression analysis between the EI and OR and EI and BIT
Compounds were tested at different concentrations ranging from 0,001 µM to 10 µM involving a total of 16,629 mosquitoes
Regression was used to derive models which were subsequently used for prediction and validation of transmission blocking behavior in the mosquito using a randomly selected set of compounds
Based on these results we further proposed a simple pathway for progression of molecules in SMFA thereby enabling the efficient use of mosquitoes and assay slots for screening molecules with TrB activity
with a Root Mean Square Error (RMSE) of 22.51%. The model provides the average OR expected in the mosquito when EI activity of a particular compound is known.
</a>ROC curves of OR classification with EI
Empirical ROC curves for OR binary classification at moving cutoffs of EI
Curves are shown for binarization cutoffs for OR of 20 (red)
we can see that when EI = 0 we obtain an average OR of 24% indicating that even if some compounds demonstrate no exflagellation inhibition
they still could show some effect on oocyst reduction (there were still some compounds showing positive OR value)
This is probably due to the nature of the data set used; compounds for SMFA were selected based on previous activity in other gametocytocidal assays thereby enriching for OR-active molecules
the prediction at high EI values of 80 and more is very good and highly accurate
This shows that molecules with an EI of 80% or more in the EIA most certainly will reduce the oocyst intensities by more than 80% in the mosquito
To assess the predictive power of the procedure and type of model used, a leave-out-one cross-validation was performed, where each data point is removed, and then a model is derived with the rest of the points, which is then used to predict the former. This approach gave a RMSE of 22.93%, similar to that obtained in Eq. <a data-track="click" data-track-label="link" data-track-action="equation anchor" href="/articles/s41598-018-26125-w#Equ4">4</a>
</a>Prediction of OR. Experimental vs. predicted OR for external set (4 compounds, 17 data points and 1747 fed mosquitoes). Predictions were obtained using Eq. <a data-track="click" data-track-label="link" data-track-action="equation anchor" href="/articles/s41598-018-26125-w#Equ4">4</a>
</a>Relationship between OR and BIT. Scatter plot of OR vs. BIT for the 40 different compounds and concentrations as described in Fig. <a data-track="click" data-track-label="link" data-track-action="figure anchor" href="/articles/s41598-018-26125-w#Fig1">1</a>
Each point is the average of up to three repetitions of SMFA with a total of 14,882 fed mosquitoes
with error bars corresponding to standard errors
no saturation occurred and therefore a straight line was fitted to the points
resulting in the following best-fit equation:
This corresponds to an RMSE of 21.07% (26.37% in leave-out-one cross-validation)
The nonlinear relationship between EI and OR is compensated by the nonlinear relationship between OR and BIT
so that the resulting EI versus BIT relationship seems to be best described by a linear model
</a>ROC curves of BIT classification with EI
Empirical ROC curves for BIT binary classification at moving cutoffs of EI
Curves are shown for binarization cutoffs for BIT of 20 (red)
</a>Prediction of BIT. Experimental vs predicted BIT for external set (4 compounds, 17 data points). Predictions were obtained with Eq. <a data-track="click" data-track-label="link" data-track-action="equation anchor" href="/articles/s41598-018-26125-w#Equ5">5</a>
A significant linear relationship is observed; the best-fit linear regression model (black line in the Figure) corresponds to equation
</a>Relationship between EI pIC50 and BIT pIC50. Scatter plot of EI pIC50 vs. BIT pIC50 for 11 compounds. Each point is the mean of up to four measurements, with error bars corresponding to the standard errors. Best-fit line (Eq. <a data-track="click" data-track-label="link" data-track-action="equation anchor" href="/articles/s41598-018-26125-w#Equ7">7</a>) is shown in black; confidence bands including both EI and BIT pIC50 errors are included in dark grey
To validate this model, the BIT pIC50s were also determined for the external set of 4 compounds (Supplementary Table <a data-track="click" data-track-label="link" data-track-action="supplementary material anchor" href="/articles/s41598-018-26125-w#MOESM3">2</a>). Table <a data-track="click" data-track-label="link" data-track-action="table anchor" href="/articles/s41598-018-26125-w#Tab3">3</a> shows the predicted and actual BIT pIC50’s
maximum difference of 0.3 pIC50 units) is observed between predicted and actual values
although in the case of BIT a trend is observed in the pIC50 of EI to under predict the pIC50’s of BIT
SMFA outcome is interdependent on gametocyte infectivity and mosquito susceptibility to P
as well as the number of mosquitoes in each treatment
A useful parameter for determining infectivity of mature gametocytes to mosquitoes after the long in vitro culturing process relies on the ability of male gametocytes to form functional motile male gametes
we not only used this assay to determine culture viability but also used it to study the inhibitory effect of compounds on male gamete formation
The same drug-treated gametocytes were used both
in EIA and for preparation of bloodmeal for mosquito feeds in SMFA
This is extremely advantageous as we could analyze the outcomes of the same parasites in two different assays enabling a direct correlation between the outcomes of an in vitro assay based on inhibition of male gamete formation
and the final output in the mosquito in terms of reduction in oocyst intensity and prevalence
We employed a data set derived from the analysis of a total of 40 compounds assayed either at a single or multiple concentrations (131 data points) using a total of 14,882 mosquitoes
This allowed us to estimate regression models for both EI vs
OR (in the form of a four-parameter logistic equation) and EI vs
BIT (in the form of a simple linear regression model) at a single concentration
The models were internally cross-validated by leave-one-out cross-validation
and externally validated with a validation set
showing good predictive powers in both cases with an error similar to that of the training set
The observed trend seems to be general and not dependent on chemical family
simple linear regression models were derived to predict pIC50’s of OR and BIT from pIC50’s of EI
which gave reasonable predictions when used with an external validation set of compounds
these models derived from either the single concentration or pIC50 data
show the efficacy of the EIA in predicting SMFA outcomes
and can be used in the prioritization of compounds to be tested in SMFAs
Besides providing essential information about new chemical structures and scaffolds to serve as starting points for drug discovery programs
results from this initial single point SMFA also help give an idea about the range of concentrations for performing a more robust full dose-response mosquito feeding assay but these are not always informative
It would be advantageous to have simple and easy to use predictive tools derived from in vitro assays (as presented here) which could estimate TrBP in the mosquito thereby pre-empting the need to perform several SMFAs in order to determine concentration range and TrB activity
the EI pIC50 will be used to predict the OR and BIT IC50’s in SMFA
we can decide on the range and number of concentrations at which to perform SMFA to determine the actual OR and BIT of the compound in the mosquito
compounds which are categorized as having poor TrBP based on the EI values
cannot be written off as having no TrBP considering that even at zero EI
we do see OR and BIT in mosquitoes for some compounds
if the EI for these compounds at a fixed concentration of 1 µM is very low
there is a low probability that these will show high OR and BIT in the SMFA
Depending on the stage of development of these compounds (late lead or pre-clinical candidate) such compounds can be progressed into SMFA at a single concentration of 1 µM to confirm activity in the mosquito
thus obviating the need to perform tedious dose dependent SMFA’s at several concentrations with large numbers of mosquitoes
this study is unique in that it uses identical parasites in two different assays and therefore shows the direct correlation between results from an in vitro male gamete formation assay and mosquito feeding assay
we have demonstrated the ability of efficiently using an in vitro generated outcome to predict TrB ability of compounds in complex in vivo biological systems
using models obtained from these analyses we proposed a simple pathway for progression of molecules in SMFA thereby contributing to efficient use of mosquitoes and slots for screening molecules with TrB in mosquitoes
Mosquito rearing facilities are located at Tres Cantos Medicines Development Campus at GSK (Madrid
Anopheles stephensi colony was established in 2013 from eggs kindly provided by Michael Delves and Mark Tunncliffe from Imperial College
Mosquitoes were maintained in climate controlled chambers (Panasonic MLR 352-H) at a temperature of 26.5 ± 1 °C
14 L:10D photoperiod and a relative humidity of 75 ± 5%
with ad libitum access to 10% glucose/water solution + 1% Karo® syrup
Gametocyte induction was initiated at 0.5% parasitemia (&gt;70% rings) and 4% hematocrit in RMPI1640 with 30 mM bicarbonate and 5 mM hypoxanthine at 50 mL final volume in T75 flasks
Media was made complete with 5% human serum A+ and 0.5% albumax (5% Albumax II-Sigma from 20X stock solution)
Cultures were maintained for up to 20 days with daily media change and without addition of fresh RBC in a gassed incubator at 37 °C
media was replaced with a serum only gametocyte treatment media (same as asexual culture media described above)
Cultures were monitored daily after day 13 using Giemsa stained smears for Stage V gametocytemia
male and female gametocyte ratio and by performing exflagellation assay for viability
Data in this study was generated using results from a total of 44 different compounds belonging to different chemical classes and were either leads, late leads, pre-clinical candidates or in clinical stages of development (Table <a data-track="click" data-track-label="link" data-track-action="table anchor" href="/articles/s41598-018-26125-w#Tab1">1</a>) and are referred to in the figures and tables in the Results section as compound 1
Compounds were prepared fresh as a stock solution of 10 mM in 100% of DMSO and added to mature gametocytes as described below
These were screened either at a single concentration or in a full dose at four or five different concentrations
In vitro cultures used for EIA and SMFA had gametocytemia between 1 to 3% Stage V
male to female ratio of not less than 1:2 and exflagellations of more than 20 centers/field at 100X total magnification
Gametocyte cultures (5 mL) were incubated for 24 hours with the required concentration of compound in the same final concentration of DMSO (0.1%) and an aliquot (100 µL) of these cultures was used in the EIA described in the method below
at 24 hours post drug-exposure a fixed volume (3 mL) of spent media was removed and fresh media was re-added accompanied by compound replenishment to obtain the required final concentration
Untreated gametocytes with the same final concentration of DMSO (0.1%) as in treated gametocytes were processed in parallel
Exflagellation Inhibition Assay (EIA) was performed prior to all SMFA’s and was initially performed by manual counting of microscopic exflagellation centers and later by capturing movement of exflagellating centers over time by video microscopy
The inhibitory effect of the compound on each of the parameters observed (exflagellation centers per field
prevalence of infection) were normalized to the respective DMSO treated control and % inhibition (y) calculated
Percentage inhibition data obtained from each individual parameter was fitted to a four-parameter logistic equation with variable slope using GraphPad Prism 6.07 for which the compound concentrations (x) were first transformed to respective log10 values
The data were fitted using these transformed values of x without defining constraints
The IC50 values were calculated from the anti-log of the x value corresponding to the 50% inhibition obtained by interpolating the XY coordinate table obtained in GraphPad Prism
pIC50’s were calculated from the IC50 as described in Results
All the statistical analyses and graphs were generated with the software R
The pROC package was employed for the ROC analyses
and the ggplot2 and base packages for generating the plots
The 95% confidence bands of the regression models (grey bands in the figures)
were obtained through Monte Carlo simulations that included both the error in EI and SMFA parameters (OR and BIT)
leave-out-one cross validations were performed
The regression models were applied to external validation sets to ascertain their predictive power in data not used to train the models (external validation)
As a measure of error the Root Mean Squared Error (RMSE) was used
All methods in this study were carried out in accordance with GlaxoSmithKline guidelines and regulations
Human biological samples used in the study have been obtained in accordance with all relevant laws including
the Human Tissue Act 2004 and the Medical Research Council (MRC) Guidelines entitled “Human Tissue and Biological Samples for use in Research” regarding the collection
Human red blood cells used in the study were obtained from Biobank of Castilla y Leon
BST and Centro de Transfusiones de Madrid and the relevant ethics committee approvals have been obtained from Autonoma University of Madrid (CEI-45-890) to enable the use these samples from human subject volunteers or other donors
All use of human biological samples in this study was in accord with the terms of the informed consents given by the sample donors
All experiments for use of mice were ethically reviewed and approved by the GlaxoSmithKline Diseases of the Developing World (DDW) Group Ethical Committee on Animal Research and were conducted according to Spanish legislation
European Directive 2010/63/EU and GlaxoSmithKline policy on the Care
Welfare and Treatment of Laboratory animals
A research agenda for malaria eradication: drugs. PLoS medicine 8, e1000402, <a href="https://doi.org/10.1371/journal.pmed.1000402" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1371/journal.pmed.1000402">https://doi.org/10.1371/journal.pmed.1000402</a> (2011)
Alonso, P. L. et al. A research agenda to underpin malaria eradication. PLoS medicine 8, e1000406, <a href="https://doi.org/10.1371/journal.pmed.1000406" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1371/journal.pmed.1000406">https://doi.org/10.1371/journal.pmed.1000406</a> (2011)
Burrows, J. N. et al. New developments in anti-malarial target candidate and product profiles. Malaria Journal 16, <a href="https://doi.org/10.1186/s12936-016-1675-x" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1186/s12936-016-1675-x">https://doi.org/10.1186/s12936-016-1675-x</a> (2017)
Birkholtz, L. M., Coetzer, T. L., Mancama, D., Leroy, D. &amp; Alano, P. Discovering New Transmission-Blocking Antimalarial Compounds: Challenges and Opportunities. Trends in Parasitology 32, 669–681, <a href="https://doi.org/10.1016/j.pt.2016.04.017" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1016/j.pt.2016.04.017">https://doi.org/10.1016/j.pt.2016.04.017</a> (2016)
Delves, M. J. et al. Male and female Plasmodium falciparum mature gametocytes show different responses to antimalarial drugs. Antimicrobial agents and chemotherapy 57, 3268–3274, <a href="https://doi.org/10.1128/AAC.00325-13" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1128/AAC.00325-13">https://doi.org/10.1128/AAC.00325-13</a> (2013)
Ruecker, A. et al. A male and female gametocyte functional viability assay to identify biologically relevant malaria transmission-blocking drugs. Antimicrobial agents and chemotherapy 58, 7292–7302, <a href="https://doi.org/10.1128/aac.03666-14" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1128/aac.03666-14">https://doi.org/10.1128/aac.03666-14</a> (2014)
Miura, K. et al. Qualification of Standard Membrane-Feeding Assay with Plasmodium falciparum Malaria and Potential Improvements for Future Assays. PloS one 8, <a href="https://doi.org/10.1371/journal.pone.0057909" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1371/journal.pone.0057909">https://doi.org/10.1371/journal.pone.0057909</a> (2013)
Li, T. et al. Robust, reproducible, industrialized, standard membrane feeding assay for assessing the transmission blocking activity of vaccines and drugs against Plasmodium falciparum. Malaria Journal 14, <a href="https://doi.org/10.1186/s12936-015-0665-8" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1186/s12936-015-0665-8">https://doi.org/10.1186/s12936-015-0665-8</a> (2015)
Stone, W. J. &amp; Bousema, T. The Standard Membrane Feeding Assay: Advances Using Bioluminescence. Methods in molecular biology (Clifton, N. J.) 1325, 101–112, <a href="https://doi.org/10.1007/978-1-4939-2815-6_9" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1007/978-1-4939-2815-6_9">https://doi.org/10.1007/978-1-4939-2815-6_9</a> (2015)
Dechering, K. J. et al. Modelling mosquito infection at natural parasite densities identifies drugs targeting EF2, PI4K or ATP4 as key candidates for interrupting malaria transmission. Scientific reports 7, 17680, <a href="https://doi.org/10.1038/s41598-017-16671-0" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1038/s41598-017-16671-0">https://doi.org/10.1038/s41598-017-16671-0</a> (2017)
The action of metabolic inhibitors on microgametogenesis in Plasmodium yoelii nigeriensis
Leba, L. J. et al. Use of Plasmodium falciparum culture-adapted field isolates for in vitro exflagellation-blocking assay. Malar J 14, 234, <a href="https://doi.org/10.1186/s12936-015-0752-x" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1186/s12936-015-0752-x">https://doi.org/10.1186/s12936-015-0752-x</a> (2015)
Lelievre, J. et al. Activity of clinically relevant antimalarial drugs on Plasmodium falciparum mature gametocytes in an ATP bioluminescence “transmission blocking” assay. PloS one 7, e35019, <a href="https://doi.org/10.1371/journal.pone.0035019" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1371/journal.pone.0035019">https://doi.org/10.1371/journal.pone.0035019</a> (2012)
Miguel-Blanco, C. et al. Imaging-based high-throughput screening assay to identify new molecules with transmission-blocking potential against Plasmodium falciparum female gamete formation. Antimicrobial agents and chemotherapy 59, 3298–3305, <a href="https://doi.org/10.1128/AAC.04684-14" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1128/AAC.04684-14">https://doi.org/10.1128/AAC.04684-14</a> (2015)
Duffy, S. &amp; Avery, V. M. Identification of inhibitors of Plasmodium falciparum gametocyte development. Malar J 12, 408, <a href="https://doi.org/10.1186/1475-2875-12-408" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1186/1475-2875-12-408">https://doi.org/10.1186/1475-2875-12-408</a> (2013)
Plouffe, D. M. et al. High-Throughput Assay and Discovery of Small Molecules that Interrupt Malaria Transmission. Cell host &amp; microbe 19, 114–126, <a href="https://doi.org/10.1016/j.chom.2015.12.001" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1016/j.chom.2015.12.001">https://doi.org/10.1016/j.chom.2015.12.001</a> (2016)
Churcher, T. S. et al. Measuring the blockade of malaria transmission - An analysis of the Standard Membrane Feeding Assay. International Journal for Parasitology 42, 1037–1044, <a href="https://doi.org/10.1016/j.ijpara.2012.09.002" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1016/j.ijpara.2012.09.002">https://doi.org/10.1016/j.ijpara.2012.09.002</a> (2012)
Heterogeneity in patterns of malarial oocyst infections in the mosquito vector
Almela, M. J. et al. A New Set of Chemical Starting Points with Plasmodium falciparum Transmission-Blocking Potential for Antimalarial Drug Discovery. PloS one 10, e0135139, <a href="https://doi.org/10.1371/journal.pone.0135139" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1371/journal.pone.0135139">https://doi.org/10.1371/journal.pone.0135139</a> (2015)
Miguel-Blanco, C. et al. Hundreds of dual-stage antimalarial molecules discovered by a functional gametocyte screen. Nature communications 8, <a href="https://doi.org/10.1038/ncomms15160" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1038/ncomms15160">https://doi.org/10.1038/ncomms15160</a> (2017)
Ghosh, A. K., Dinglasan, R. R., Ikadai, H. &amp; Jacobs-Lorena, M. An improved method for the in vitro differentiation of Plasmodium falciparum gametocytes into ookinetes. Malar J 9, 194, <a href="https://doi.org/10.1186/1475-2875-9-194" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1186/1475-2875-9-194">https://doi.org/10.1186/1475-2875-9-194</a> (2010)
NIH Image to ImageJ: 25 years of image analysis
Kamentsky, L. et al. Improved structure, function and compatibility for CellProfiler: modular high-throughput image analysis software. Bioinformatics (Oxford, England) 27, 1179–1180, <a href="https://doi.org/10.1093/bioinformatics/btr095" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.1093/bioinformatics/btr095">https://doi.org/10.1093/bioinformatics/btr095</a> (2011)
<a data-track="click" data-track-action="download citation references" data-track-label="link" rel="nofollow" href="https://citation-needed.springer.com/v2/references/10.1038/s41598-018-26125-w?format=refman&amp;flavour=references">Download references</a>
This work was supported by the Medicines for Malaria Venture
Esperanza Herreros and Antonio Martinez-Escandell for their immense contributions towards establishing the Insectary at GSK
Francisco Javier Gamo (GSK) is kindly acknowledged for critically reading the manuscript
BST and Centro de Transfusiones de Madrid for providing human red blood cells used in the study
Present address: Biostatistics and Bioinformatics Unit
Present address: Complejo Asistencial Universitario de León
Jesús-Luís Presa Matilla &amp; Janneth Rodrigues
micro-dissections and microscopic enumeration
mosquito colony maintenance and supply and in vitro gametocyte culture
design and interpretation and all figures in the manuscript were done by G.C.
developed scripts for the modified exflagellation method and data analysis
conceived and designed the study and analyzed EIA and SMFA data
contributed to study design and provided compounds and compound information
All authors reviewed and approved the manuscript
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations
<a data-test="citation-link" data-track="click" data-track-action="download article citation" data-track-label="link" data-track-external="" rel="nofollow" href="https://citation-needed.springer.com/v2/references/10.1038/s41598-018-26125-w?format=refman&amp;flavour=citation">Download citation</a>
DOI: https://doi.org/10.1038/s41598-018-26125-w
Sign up for the Nature Briefing: Translational Research newsletter — top stories in biotechnology