低深度测序数据的基因型填充优化与回归模型性能分析

打开文本图片集

中图分类号：S332 文献标识码：A文章编号：0439-8114（2025）07-0203-04DOI：10.14088/j.cnki.issn0439-8114.2025.07.035开放科学（资源服务）标识码（OSID）：

Optimization of genotype imputation for low-depth sequencing data and performance analysis of regression models

XIANG Chong，CHEN Can

（School of Dataand Information，ChangjiangPolytechnic，Wuhan 43oo7O，China）

Abstract：Anewmethodsuitableforanalyzinglow-depth sequencing genomicdatawasestablishedbyoptimizing genotypeimputationalgorithmsandscreningoptialregressionmodels.Thesultsshowedthatcomparedtotepreoptiizationalgorit，thac racy of the optimized genotype imputation algorithm increased from 95% to 98% . Meanwhile，parameter tuning and efficient algorithm selectionreducedthesingleimputationtimefrom24hoursto12hours，significantlyimprovingprocesingeficiency.Forcotiuous phenotypicanalysis（e.g，quantitativetraitsinGWAS），theridgeregresionmodelandlinearregressionmodelperforedwllAt 1.0Xsequencingdepth，theirMSEswereO.O7andO.08，andAcuracieswereO.82and.80，respectively.Whenandlingcaiication problems（e.g.，genomicselection），eLogisticegressnodeldmonstratedsinificantadvantagesduetoitsprobabilisticodeling characteristics.This model showed good Classification performance（ AUC =0.90），significantly outperforming theLinear regression model （AUC=0.85）.

Keywords：low-epthsequencingdata;genotypeimputation;ridgeregresionmodels；performanceanalysis；linearregresionmod el; Logistic regression model

随着基因组学研究的不断深人，高通量测序技术已成为解析生物遗传信息的重要手段。（剩余5073字）