Dataset Information

Computing the relative binding affinity of ligands based on a pairwise binding comparison network.

ABSTRACT: Structure-based lead optimization is an open challenge in drug discovery, which is still largely driven by hypotheses and depends on the experience of medicinal chemists. Here we propose a pairwise binding comparison network (PBCNet) based on a physics-informed graph attention mechanism, specifically tailored for ranking the relative binding affinity among congeneric ligands. Benchmarking on two held-out sets (provided by Schrödinger and Merck) containing over 460 ligands and 16 targets, PBCNet demonstrated substantial advantages in terms of both prediction accuracy and computational efficiency. Equipped with a fine-tuning operation, the performance of PBCNet reaches that of Schrödinger's FEP+, which is much more computationally intensive and requires substantial expert intervention. A further simulation-based experiment showed that active learning-optimized PBCNet may accelerate lead optimization campaigns by 473%. Finally, for the convenience of users, a web service for PBCNet is established to facilitate complex relative binding affinity prediction through an easy-to-operate graphical interface.

SUBMITTER: Yu J

PROVIDER: S-EPMC10766524 | biostudies-literature | 2023 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Computing the relative binding affinity of ligands based on a pairwise binding comparison network.

Yu Jie J Li Zhaojun Z Chen Geng G Kong Xiangtai X Hu Jie J Wang Dingyan D Cao Duanhua D Li Yanbei Y Huo Ruifeng R Wang Gang G Liu Xiaohong X Jiang Hualiang H Li Xutong X Luo Xiaomin X Zheng Mingyue M

Nature computational science 20231019 10

Structure-based lead optimization is an open challenge in drug discovery, which is still largely driven by hypotheses and depends on the experience of medicinal chemists. Here we propose a pairwise binding comparison network (PBCNet) based on a physics-informed graph attention mechanism, specifically tailored for ranking the relative binding affinity among congeneric ligands. Benchmarking on two held-out sets (provided by Schrödinger and Merck) containing over 460 ligands and 16 targets, PBCNet ...[more]

PMID: 38177766

Similar Datasets

Project description:BackgroundPregnane X receptor (PXR) agonists inhibit liver fibrosis. However, the rodent PXR activator pregnenolone 16alpha carbonitrile (PCN) blocks, in vitro, hepatic stellate cell-to-myofibroblast trans-differentiation and proliferation in cells from mice with a disrupted PXR gene, suggesting there is an additional anti-fibrogenic drug target for PCN. The role of the low affinity glucocorticoid binding site (LAGS) - which may be identical or associated with the progesterone receptor membrane component 1 (PGRMC1) - in mediating this anti-fibrogenic effect has been examined, since binding of dexamethasone to the LAGS in liver microsomal membranes has previously been shown to be inhibited by PCN.ResultsQuiescent rat and human hepatic stellate cells (HSC) were isolated from livers and cultured to generate liver myofibroblasts. HSC and myofibroblasts expressed PGRMC1 as determined by RT-PCR and Western blotting. Quiescent rat HSC also expressed the truncated HC5 variant of rPGRMC1. Rat PGRMC1 was cloned and expression in COS-7 cells gave rise to specific binding of radiolabelled dexamethasone in cell extracts that was inhibited by PCN, suggesting that PGRMC1 may be identical to LAGS or activates LAGS binding activity. Liver microsomes were used to screen a range of structurally related compounds for their ability to inhibit radiolabelled dexamethasone binding to rat LAGS. These compounds were also screened for their ability to activate rat and human PXR and to inhibit rat HSC-to-myofibroblast trans-differentiation/proliferation. A compound (4 androstene-3-one 17beta-carboxylic acid methyl ester) was identified which bound rat LAGS with high affinity and inhibited both rat and human HSC trans-differentiation/proliferation to fibrogenic myofibroblasts without showing evidence of rat or human PXR agonism. However, despite potent anti-fibrogenic effects in vitro, this compound did not modulate liver fibrosis severity in a rat model of liver fibrosis. Immunohistochemical analysis showed that rat liver myofibroblasts in vivo did not express rPGRMC1.ConclusionLAGS ligands inhibit HSC trans-differentiation and proliferation in vitro but show little efficacy in inhibiting liver fibrosis, in vivo. The reason(s) for this disparity is/are likely associated with an altered myofibroblast phenotype, in vitro, with expression of rPGMRC1 in vitro but not in vivo. These data emphasize the limitations of in vitro-derived myofibroblasts for predicting their activity in vivo, in studies of fibrogenesis. The data also demonstrate that the anti-fibrogenic effects of PCN in vivo are likely mediated entirely via the PXR.

Dataset Information

Computing the relative binding affinity of ligands based on a pairwise binding comparison network.

Publications

Computing the relative binding affinity of ligands based on a pairwise binding comparison network.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets