Widespread And Polymorphous Noncoding Amino Acids In Human Spermoblasts

Proteins are usually determined by translation of the coding sequence of the genome.However,due to post-translational modification,amino acid substitution and other reasons,their amino acid residues are rarely directly determined by the genome,the actual situation of amino acid residues tend to change,thereby changing the protein structure and affect the protein function.However,the amino acid residues in the organism are rarely determined directly in proteomics,mainly because amino acid residues that are different from the amino acids encoded are usually ignored by the general search algorithm,and secondly because the protein sequencing technique usually depends on the theory Translated on the protein database.However,by assuming that one or more undefined non-coding amino acid residues are present in the peptide break sequence,it becomes a breakthrough point for resolving those that can not match the peptide spectrum.In earlier methods,partial peptide sequences were derived from mismatched spectra that could be used as tags to search for theoretically genomic translation of the protein database,and the search results would result in unexpected post-translational modifications and amino acid substitutions.Later,non-limiting search algorithms were used to identify non-coding amino acid residues without knowing whether they were present.The mass-tolerant method was originally used to detect known modifications by allowing the difference in mass between the precursor and its fragments,and the method has recently been improved by matching a wide mass-tolerant to match a wide range of poor quality or undefined modifications The peptide sequence,found many modifications.But the main problem with these methods is still higher false positives,lower sensitivity and long search times.Here we systematically study all possible amino acid residues in human sperm cells,whose relative molecular mass is different from the amino acids encoded in the genomic sequence,called noncoding amino acids(ncAA).By measuring the difference in quality between the encoded amino acid and the actual protein residue,more than one million amino acids with a non-zero quality difference,i.e.,a side chain,were found to be altered.Then,based on these mass differences,Gaussian mixture distribution analysis and iterative regression analysis were used to determine the 424 high confidence clustering Gaussian clusters.The decision tree was established by machine learning algorithm.849 highly reliable ncAs were identified and distributed in 35,274 Protein sites.Among them,180 kinds of mass difference clusters were found to have a non-reported amino acid side chain structure;105 kinds of ncAAs were matched to the type of amino acid substitutions,of which 40 were confirmed by transcriptome sequencing.In addition,according to ANOVA analysis,some ncAAs were found to be specific in the normal population,suggesting that these ncAAs may be related to population differences.And some ncAAs are distributed in severe infirm and normal populations,suggesting that these ncAAs are associated with the pathogenesis,and some of the phosphorylation sites have been reported by previous studies.Our study shows that ncAAs are abundantly present in sperm cells,mainly because of nucleotide polymorphisms,post-translational modifications,and some unknown mechanisms that are important for the diagnosis of disease and drug targeted therapy.