Genome-wide Association Study And Genomic Prediction For Growth Traits In Largemouth Bass (Micropterus Salmoides)

The largemouth bass(Micropterus salmoides)is an important freshwater aquaculture fish in China,and its production has shown a trend of growth in recent years.As the large-scale breeding of largemouth bass continues to increase,genetic improvement in growth,body size,disease resistance,and other aspects will be of great significance to its industry.This study collected data on seven growth-related traits,including weight,length,height,thickness,tail stem height,tail stem length,and plumpness,from two batches of 1,066 largemouth bass individuals.The researchers used second-generation sequencing to mine SNPs from all individuals and evaluated the fish’s growth traits using whole-genome association analysis and genomic prediction.The main results are as follows:1.Descriptive statistical analysis of the phenotype data from the two batches of largemouth bass,and correlation tests on the two batches of data,showed that the average body weight of the second batch of largemouth bass was higher than that of the first batch due to the more mature development of the ovaries in the female fish.We also found a significant correlation between body weight,body height,and caudal peduncle height in largemouth bass.2.This study performed whole-genome association analysis on the 1,066 largemouth bass individuals and identified 16 SNP loci associated with 7 growth traits after filtering,located on chromosomes 1,3,13,14,16,18,19,21,and 23.The results also identified 34 candidate genes associated with growth traits.Among them,the genes around the SNP locus on chromosome 23 were largely consistent with the GWAS results for body weight,body height,and caudal peduncle height,including rerg,jup,cd63,etv6,lrig3,sox5,myf6,myf5,and igf1.3.The study selected myf5 and igf1 from the growth-related candidate genes and constructed tissue expression profiles using q RT-PCR.The results showed that the myf5 and igf1 genes and SNP loci used in the study were indeed associated with largemouth bass growth4.Four models were used to evaluate the accuracy of genomic prediction for largemouth bass,and the results showed that LASSO performed poorly,while ridge regression and SVR(Linear)had similar prediction accuracies.The SVR(Poly)model had better prediction accuracy for the seven growth traits of largemouth bass,with a prediction accuracy of up to 60% for height and tail stem length when using 75,433 SNPs for whole-genome prediction.5.This study extracted the GWAS results for largemouth bass and divided them into nine genotype files with different numbers of SNPs.They also used the four models to evaluate the accuracy of genomic prediction for growth traits.The results showed that the prediction accuracy of the LASSO model was the worst.Using 500 SNPs in ridge regression,and using 1,000 SNPs in SVR(linear),had higher respectively prediction accuracy for body length than using 75,433 SNPs,with increases of 8.56% and 4.01%.The SVR(Poly)model had the best prediction accuracy for most growth traits of largemouth bass,with better model fitting performance.Moreover,the prediction accuracy of the model increased with the number of SNPs used.This study conducted a GWAS on the growth traits of large-mouth bass,identifying candidate gene loci related to growth traits and evaluating the accuracy of genome prediction.The results showed that the SVR(Poly)model had the best performance in predicting the genome of large-mouth bass.This study lays an important foundation for the genetic breeding of growth traits in large-mouth bass.