Grant: Master's Award: Canada Graduate Scholarships
University of British Columbia
$17,500.00
Development, Advancement and Improvement of Computational Drug Discovery Methods for Prostate Cancer
EF 2023-24
Résumé
This research is aimed to develop DD-Light, an innovative ultra-large docking tool. Stemming from the Deep Docking (DD) platform developed in response to the COVID-19 pandemic, DD-Light employs active deep learning and a 2D fingerprint ligand representation for accelerated drug discovery against SARS-CoV-2.
The goal of DD-Light development is determining speed increases and acceptable accuracy trade-offs for pre-trained docking emulators deployed as web servers. Using multiprocessing, multi-GPU with PyTorch, and a revamped sampling mechanism, the approach will archieve faster inference times, preventing resource exhaustion.
An extensive evaluation of molecular descriptors, including QSAR, topological descriptors, and 3D-MoRSE, is planned. Solutions, offering up to thousands-fold acceleration in A.I.-accelerated hit delivery, will be shared with the Canadian scientific community as plug-ins for popular CADD protocols, aligning with the proposal's aim to create an open-source web tool for ultra-large docking.
Adhering to Open Science principles, the entire DD-Light codebase will be released on GitHub. This enables scientists to download and install the optimized protocol for billion-molecule docking campaigns, with the proposal aiming to reduce the timeline for generating docking scores from 2-4 weeks to minutes, fostering experimentation and innovation in drug discovery.
In conclusion, DD-Light advances A.I.-accelerated drug discovery and equips global scientists with an efficient tool to navigate the Chemical Universe, expediting the search for vital small molecule therapeutics.
Programme
Master's Award: Canada Graduate Scholarships
Bénéficiaire
Emplacement
British Columbia, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
CIHR Funding Decisions Database
Tel que publié
Id
506352
Name
Manskaia Ekaterina
Region
British Columbia
Country
Canada
Orgtype
Institution Paid; Research Institution
Prcname2
Special Cases - Awards Programs
Projecttitle
Development, Advancement and Improvement of Computational Drug Discovery Methods for Prostate Cancer
Programtype2
Studentship Programs
Orgnameinp2
University of British Columbia
Cihramount2
$17,500
Cihrequipment2
$0
Programname2
Master's Award: Canada Graduate Scholarships
Keyworddelim
Cancer Treatment; Computational Drug Discovery; Covid-19; Deep Learning Acceleration; Drug Development; Drug Discovery Acceleration; Health Research; In Silico; Machine Learning; Prostate Cancer
Abstract
This research is aimed to develop DD-Light, an innovative ultra-large docking tool. Stemming from the Deep Docking (DD) platform developed in response to the COVID-19 pandemic, DD-Light employs active deep learning and a 2D fingerprint ligand representation for accelerated drug discovery against SARS-CoV-2.
The goal of DD-Light development is determining speed increases and acceptable accuracy trade-offs for pre-trained docking emulators deployed as web servers. Using multiprocessing, multi-GPU with PyTorch, and a revamped sampling mechanism, the approach will archieve faster inference times, preventing resource exhaustion.
An extensive evaluation of molecular descriptors, including QSAR, topological descriptors, and 3D-MoRSE, is planned. Solutions, offering up to thousands-fold acceleration in A.I.-accelerated hit delivery, will be shared with the Canadian scientific community as plug-ins for popular CADD protocols, aligning with the proposal's aim to create an open-source web tool for ultra-large docking.
Adhering to Open Science principles, the entire DD-Light codebase will be released on GitHub. This enables scientists to download and install the optimized protocol for billion-molecule docking campaigns, with the proposal aiming to reduce the timeline for generating docking scores from 2-4 weeks to minutes, fostering experimentation and innovation in drug discovery.
In conclusion, DD-Light advances A.I.-accelerated drug discovery and equips global scientists with an efficient tool to navigate the Chemical Universe, expediting the search for vital small molecule therapeutics.
Competitiondate
202312
Approvedterm2
1 yr 0 mth
Pinamesdelim
Manskaia, Ekaterina
Approvedterm
1 an 0 mois
Cihrcontribution2
~ $17,500~ $0~