RESUMO
Fragment-based drug discovery is a widely used strategy for drug design in both academic and pharmaceutical industries. Although fragments can be linked to generate candidate compounds by the latest deep generative models, generating linkers with specified attributes remains underdeveloped. In this study, we presented a novel framework, DRlinker, to control fragment linking toward compounds with given attributes through reinforcement learning. The method has been shown to be effective for many tasks from controlling the linker length and log P, optimizing predicted bioactivity of compounds, to various multiobjective tasks. Specifically, our model successfully generated 91.0% and 93.9% of compounds complying with the desired linker length and log P and improved the 7.5 pChEMBL value in bioactivity optimization. Finally, a quasi-scaffold-hopping study revealed that DRlinker could generate nearly 30% molecules with high 3D similarity but low 2D similarity to the lead inhibitor, demonstrating the benefits and applicability of DRlinker in actual fragment-based drug design.
Assuntos
Desenho de Fármacos , Descoberta de DrogasRESUMO
INTRODUCTION: The present study was conducted to determine the association of transforming growth factor-beta (TGF-ß) gene polymorphism and myopia. METHOD: Four hundred twelve articles were identified, of which 11 articles with 5213 participants in 4 countries were included in the final analysis. Review Manager software (RevMan, version 5.4) was used for data analysis. RESULT: Odds ratio (OR) value of TGF-ß1 rs1800469 is 1.33 (95% confidence interval [CI] = 1.15-1.54) in the allelic model; in the dominant model is 1.76 (95% CI = 1.16-2.67); in homozygous model is 5.98 (95% CI = 4.31-8.06). OR value of TGF-ß1 rs4803455 is 0.62 (95% CI = 0.43-0.88) in recessive model. TGF-ß2 is not associated with myopia. Relevant study on TGF-ß3 is scarce. CONCLUSION: Our systematic review and meta-analysis found that TGF-ß1 rs4803455 and rs1800469 were correlated with myopia.