首页|Data on Machine Learning Reported by Valerio Tazzari and Colleagues ("DompeKeys" a set of novel substructure-based descriptors for efficient chemical space mappi ng,development and structural interpretation of machine learning models,and ...)
Data on Machine Learning Reported by Valerio Tazzari and Colleagues ("DompeKeys" a set of novel substructure-based descriptors for efficient chemical space mappi ng,development and structural interpretation of machine learning models,and ...)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-New research on Machine Learning is th e subject of a report.According to news reporting out of Naples,Italy,by News Rx editors,research stated,"The conversion of chemical structures into compute r-readable descriptors,able to capture key structural aspects,is of pivotal im portance in the field of cheminformatics and computer-aided drug design.Molecul ar fingerprints represent a widely employed class of descriptors; however,their generation process is time-consuming for large databases and is susceptible to bias." Our news journalists obtained a quote from the research,"Therefore,descriptors able to accurately detect predefined structural fragments and devoid of lengthy generation procedures would be highly desirable.To meet additional needs,such descriptors should also be interpretable by medicinal chemists,and suitable fo r indexing databases with trillions of compounds.To this end,we developed-as i ntegral part of EXSCALATE,Domp?'s end-to-end drug discovery platform-the DompeK eys (DK),a new substructurebased descriptor set,which encodes the chemical fe atures that characterize compounds of pharmaceutical interest.DK represent an e xhaustive collection of curated SMARTS strings,defining chemical features at di fferent levels of complexity,from specific functional groups and structural pat terns to simpler pharmacophoric points,corresponding to a network of hierarchic ally interconnected substructures.Because of their extended and hierarchical st ructure,DK can be used,with good performance,in different kinds of applicatio ns.In particular,we demonstrate how they are very well suited for effective ma pping of chemical space,as well as substructure search and virtual screening.N otably,the incorporation of DK yields highly performing machine learning models for the prediction of both compounds' activity and metabolic reaction occurrenc e."
NETL
NSTL
