Upgrading Of Anti-CPRISPRdb And Clustering Analysis Of Acr Proteins

The Anti-CRISPRdb released by our group before contains Anti-CRISPR protein(Acr)and its related information.Since its released,it has been widely used in the fileds of Acr and CRISPR(Clustered Regularly Interspaced Short Palindromic Repeats),such as constructing benchmark datasets for the identification issue,studying the relationship between ICE(integrative and conjugative element)and Acr,screening new Acr in another species,studying the evolution of Acr protein and the diversity of Acr family,and studying the effect of Acr on the horizontal transfer of resistance genes,etc.At present,the newly discovered Acr type,family,inhibitory mechanism,inhibitory strength and other information are accumulated as time going on.Recent studies have shown that the resistance genes on mobile genetic elements(MGEs)that against prokaryotic immune systens are usually located close to each other.Based on this,it is reasonably speculated that Acr's neighbors have the potential to become Acr.On this basis,this work has made a major update to the database,including the update of contents and functions,mainly in the following several aspects:(1)The newly discovered Acr in recent years has been updated,the entries have been expanded to 3700,and the family has been expanded to nearly 100;(2)The new version contains information of new identified inhibitory mechanism,inhibitory strength,and inhibitory stage;(3)We also Added evaluation of Acr's neighboring proteins in the database,and added Acr candidates by considering Acrs neighbours and PSI-BLAST search.In order to solve the time-consuming issue of MGE recognition and study the origin of Acr,this work proposes the parameter CUBRank.Based on the analysis of our updated database and our proposed CUBRank,this work finds that there is a significant difference between CUBRank and Ac Ranker.CUBRank can complement with Ac Ranker regarding Acr prediction.Meanwhile,we used it to conduct a investigation of Acr origination,and we found that Acr may originate from the intergenic region of the viral genome supporting that Acr was born through the de novo mechanism,and CUBRank can be used to measure the possibility of genes appearing in genomic islands,which can become a new parameter for the discovery of Acr located on such hot spots,and to measure which fragments are likely to be genomic islands or prophage fragments.In addition,this work also conducts a simple analysis of the cluster tendency of Acr,because we found that there may be more than one acr gene in a related locus,we determined that acr does have a clustering tendency.We therefore employed it to improve the precision of our previously developed software Acr Detector.Although the accuracy of Acr Detector did not improve,the clustering tendency of Acr also supported the speculation that the neighbor proteins of Acr were selected as candidates in the first part of this work.And provide a theoretical basis for estimating neighbor proteins in the database.The clustering analysis of Acr is based on the existing Acr that have been discovered,and those undetected proteins bearing Acr function do not considered in this analysis,so this part of the Acrs clustering analysis has a statistical bias more or less.