Design And Implementation Of Kinship Prediction System

Currently,most of the kinship identifications in China are based on autosomal STR(Short Tandem Repeat,STR)genetic markers loci.In the field of criminal investigation,Y-STR can be used to carry out family identification and assist paternity identification to provide clues for case investigation.However,Y-STR can only be used for paternal relatives,and it can't investigate maternal relatives.In recent years,the demand for complex kinship identification such as half sibling,uncle and nephew,aunt and nephew has been increasing,the existing methods are far from meeting the needs of material evidence identification.With the development of the genome-wide sequencing technology,kinship prediction technology using genome-wide SNP(Single Nucleotide Polymorphism,SNP)data has emerged and developed rapidly at the historic moment,which solved the current problem.This technology aims to give the genomewide data file of all individuals in a custom population,and use the kinship prediction algorithm to calculate all existing kinship pairs in the population,and then search for the kinship between unknown individual and multiple known individuals,so as to provide strong clues for material evidence identification.However,the original prediction process of this technology is very complicated.It needs to preprocess genome-wide data,use Plink software to transform the file into binary,and then use KING software to calculate the kinship degree;if you need to know the actual family relationship,you need to reverse restore according to the kinship degree judgment method.Therefore,this paper proposes and develops a highly interactive analysis system-the Kinship Prediction System,its processing flow is interfaced,automated and simplified,it can preprocess the genome-wide data and draw the final results as a family tree automatically.The specific contents of this paper are as follows:(1)To research the algorithm for unifying data input format,the system need to support the genome-wide sequencing data of multiple detection platforms and data formats.The staff only need to input the detected original file into the system,there is no need to write additional scripts for data format processing.(2)To research the asynchronous task processing mechanism,the system need to manage large computing tasks with message queues,and the execution status of each task can be viewed in the task list in real time.(3)To research the transformation mechanism between the kinship degree and the family tree,the system need to automatically draw a pair of kinships as a family tree according to the kinship degree judgment method for users to view directly.This system has greatly simplified the process of kinship prediction,it doesn't require operators to master the data processing script and bioinformatics software,significantly reduces the operation difficulty,improves the analysis efficiency,provides better services for DNA identification,and has important practical values in the field of human genetics and forensic genetics.