notesum.ai

Published at November 20

DrugGen: Advancing Drug Discovery with Large Language Models and Reinforcement Learning Feedback

q-bio.QM

cs.AI

Released Date: November 20, 2024

Authors: Mahsa Sheikholeslami¹, Navid Mazrouei¹, Yousof Gheisari¹, Afshin Fasihi², Matin Irajpour³, Ali Motahharynia⁴

Aff.: ¹Regenerative Medicine Research Center, Isfahan University of Medical Sciences, Isfahan, Iran; ²Department of Medicinal Chemistry, School of Pharmacy, Isfahan University of Medical Sciences, Isfahan, Iran; ³Isfahan Cardiovascular Research Center, Cardiovascular Research Institute, Isfahan University of Medical Sciences, Isfahan, Iran; ⁴Isfahan Neuroscience Research Center, Isfahan University of Medical Sciences, Isfahan, Iran

Arxiv: http://arxiv.org/abs/2411.14157v1

Targets	DrugGPT	DrugGen	U statistics	P
ACE	5.71 [5.10-6.71]	8.43 [6.65-9.06]	1475	$<10^{-16}$ *
PPARG	6.32 [5.75-6.74]	7.39 [6.61-7.95]	2208	$<10^{-10}$ *
GALM	6.12 [4.96-6.73]	6.92 [6.04-7.73]	2767	$<10^{-6}$ *
FB5L3	5.35 [4.73-6.38]	6.94 [6.36-7.48]	1723	$<10^{-14}$ *
OPSB	5.43 [4.26-6.14]	7.62 [6.84-8.07]	842	$<10^{-22}$ *
NAMPT	5.84 [5.16-6.75]	7.00 [6.19-7.70]	2616	$<10^{-7}$ *
PGK2	5.23 [4.62-6.22]	7.34 [6.36-8.52]	1212	$<10^{-18}$ *
FABP5	6.30 [5.35-7.18]	6.69 [5.80-7.60]	4382	1