Predicting Drug Response with Multi-Task Gradient-Boosted Trees in Epilepsy
bioRxiv (Cold Spring Harbor Laboratory)
Abstract
Abstract Motivation Despite the availability of numerous anti-seizure medications (ASMs), drug resistance remains a major issue for people with epilepsy. The probability of achieving seizure freedom diminishes with each unsuccessful drug trial, and the impact of genetic and clinical markers on ASM response remains unclear. To address this issue, we used state-of-the-art machine learning (ML) methods to predict the response of people with epilepsy to individual ASMs based on their clinical and genomic information. Results To overcome data sparsity for less common drugs, we implement a multi-task (MT) learning approach for gradient-boosted trees (GBTs), assuming that predicting responses to different ASMs involves similar tasks. This strategy allows models for less prevalent drugs to leverage the more abundant data available for other drugs during training. The proposed model outperforms individual and combined drug-response predictions for most drugs. Our findings identify key genomic and clinical features influencing drug response, enhancing understanding of individual drug responses in people with epilepsy, and aiding clinicians in making informed treatment decisions. Availability and Implementation Due to privacy reasons data is not publically available. The code will be made available upon acceptance under https://github.com/pfeiferAI/MT-GBT
Authors 17
-
University of Tübingen · Hertie Institute for Clinical Brain Research
Affiliation as printed
Department of Neurology and Epileptology , Hertie Institute for Clinical Brain Research , University of Tübingen , Tübingen , Germany,
University of Tuebingen
-
Luxembourg Centre for Systems Biomedicine
Affiliation as printed
Bioinformatics core, Luxembourg Centre for Systems Biomedicine (LCSB);
Bioinformatics core, Luxembourg Centre for Systems Biomedicine (LCSB)
-
Affiliation as printed
University of Luxembourg;
University of Luxembourg
-
Stefan Wolking Aachen
Affiliation as printed
RTWH Uniklinik Aachen;
RTWH Uniklinik Aachen
-
Royal College of Surgeons in Ireland
Affiliation as printed
Royal College of Surgeons in Ireland, Dublin;
Royal College of Surgeons in Ireland, Dublin
-
Royal College of Surgeons in Ireland
Affiliation as printed
Royal College of Surgeons in Ireland;
Royal College of Surgeons in Ireland
-
Belfast Health and Social Care Trust · Royal Victoria Hospital
Affiliation as printed
Royal Victoria Hospital, Belfast Health and Social Care Trust, Belfast;
Royal Victoria Hospital, Belfast Health and Social Care Trust, Belfast
-
Affiliation as printed
Hopital Universitaire de Bruxelles, Brussels;
Hopital Universitaire de Bruxelles, Brussels
-
University Medical Center Utrecht
Affiliation as printed
University Medical Center Utrecht, Utrecht;
University Medical Center Utrecht, Utrecht
-
Affiliation as printed
University of Liverpool;
University of Liverpool
-
Monash University · The Alfred Hospital
Affiliation as printed
The Alfred Hospital, Monash University, Melbourne
-
University College London · UCL Queen Square Institute of Neurology
Affiliation as printed
UCL Queen Square Institute of Neurology, London WC1N 3BG & Chalfont Centre for Epilepsy, Chalfont St Peter SL9 ORJ, United Kingdom;
UCL Queen Square Institute of Neurology, London WC1N 3BG & Chalfont Centre for Epilepsy, Chalfont St Peter SL9 ORJ, United Kingdom
-
Affiliation as printed
University of Glasgow;
University of Glasgow
-
Affiliation as printed
Rehabilitation, Ophthalmology, Genetics, Maternal and Child Health, University of Genova;
Rehabilitation, Ophthalmology, Genetics, Maternal and Child Health, University of Genova
-
Affiliation as printed
deCODE Genetics, Reykjavik;
deCODE genetics
-
University of Tübingen · Universitätsklinikum Tübingen
Affiliation as printed
Universitaetsklinikum Tuebingen;
University of Tübingen: Eberhard Karls Universitat Tubingen
-
Affiliation as printed
University of Tübingen: Eberhard Karls Universitat Tubingen
Cited by 0 stored of 0
No patents citing this paper on Lens.org (checked 2026-10-06).
References 28
-
W2030655611details pending0citations
-
W2137976898details pending0citations
-
W4283075077details pending0citations
-
W1974809694details pending0citations
-
W4239943352details pending0citations
-
W2017455791details pending0citations
-
W2040073841details pending0citations
-
W2046966869details pending0citations
-
W2061458168details pending0citations
-
W2064064788details pending0citations
-
W2100549317details pending0citations