Construction And Evaluation Of Druggable Target Discovery Methods Based On Target Identification

Development of the pharmaceutical industry is essentially to the health of human beings,new drug discovery plays a vital role in the development of human society as the critical driving force of pharmaceutical industry.In recent years,with the advances of systems biology,bioinformatics,network pharmacology,high-throughput screening,group learning and other disciplines and technologies,phenotypic drug discovery replaced by reverse pharmacology(also known as target-based drug discovery).After more than 10 years of development,use high-throughput screening of large compounds libraries against isolated biological targets which are hypothesized related to the development of the disease,a large number of drugs,especially target therapy drugs have been approved by FDA.However,drug resistance problem gradually emerged with the large number of drugs has been marked,and pharmaceutical companies developed different research strategies for drug resistance,such as drug combinations and multi-target drugs.Collect drugs,targets,the targets in the disease tissue distribution and drug resistance mutation,drug combination,targets combinations,such as information not only helps to discover new drug targets,still can have specific mutations of target screened for overcome resistance mutation compounds.Based on the above problems,this topic first about drugs,targets,drug resistance mutations and target tissue distribution map,drug combination,combination of targets,such as the comprehensive collection,increase the corresponding function module for providing database.(i)2 000 drug resistance mutations in 83 targets and 104 target/drug regulatory genes,which are resistant to 228 drugs targeting 63 diseases(49 targets of 61 drugs with patient prevalence data);(ii)differential expression profiles of 758 targets in the disease-relevant drug-targeted tissue of 12 615 patients of 70 diseases;(iii)expression profiles of 629 targets in the non-targeted tissues of 2 565 healthy individuals;(iv)1 008 target combinations of 1 764 drugs and the 1604 target combination of 664 multi-target drugs;(v)additional 48 successful,398 clinical trial and 21 research targets,473 approved,812 clinical trial and 1 120 experimental drugs.The function of proteins is essential for finding new drug targets and screening active compounds,through experimental method to discovery protein function is time-consuming and expensive,a large number of prediction of protein function algorithm such as BLAST,machine learning is the development and application.A variety of computer prediction methods and generated mass data promote the study of protein function prediction,in spite of this,however,there are a large number of proteins can neither be confirmed nor method by computer prediction method annotation.Part of the reason is the false positive and prediction accuracy of various computer prediction algorithms,so it is necessary to make a comprehensive evaluation of various algorithms.