構建臺灣人群圖參考基因組應用於HLA等位基因型判讀及評估HLA分型工具在CYP基因分型的表現

Abstract

人類白血球抗原（HLA）是人類基因組中相對複雜的區域之一，位於人類6號染色體的短臂上（chr 6p21）。HLA基因型的確定對於各種臨床應用至關重要，包括器官移植、疾病診斷和藥物過敏反應。從全基因定序獲得的短序列資料中準確解讀每個基因的兩種等位基因型對於個體是一項具有挑戰性的任務。CYP基因是一組編碼細胞色素P450（Cytochrome P450）酶的基因家族，這些酶在人體內主要參與藥物代謝以及內源性物質的代謝。CYP是人體最重要的藥物代謝系統之一，約75%的臨床藥物都是由這些酶代謝。現有的HLA分型工具，在臺灣人個體的正確性仍有上升空間，透過加入臺灣人變異點資訊，或許能這些個體在回貼時貼得更好，擷取後的序列品質上升，進而影響臺灣人在HLA等位基因判讀的正確性。本研究中旨在利用HISAT2和HISAT-genotype工具篩選自1,492名臺灣個體的全基因定序資料中提取的61,424,216個變異點，將其納入圖基因組，基於此圖基因組，本研究將擴展到HLA等位基因型的判讀。透過HISAT2工具，生成判讀來自Taiwan biobank的884個個體全基因定序資料的所需索引。本研究的目標是評估相對於以前的研究，7種等位基因的HLA基因型判讀正確率是否會在加入臺灣人變異點資訊後而有所提升。實驗結果顯示，除了HLA-DQA1有些微上升之外，其餘6個等位基因都有所下降。同時，也將評估現有的HLA分型工具，T1K和HISAT-genotype，運用於判讀44組HPRC個體全基因定序資料的CYP等位基因，並且與現有的CYP分型工具，Aldy和DRAGEN進行準確率比較。這些HLA分型工具在CYP等位基因判讀上，仍有提升空間。 本研究旨在增進對於臺灣人口HLA等位基因型的判讀正確率，為各種臨床應用和遺傳研究做出貢獻，並探索是否有更多的工具能有效的處理CYP基因分型的問題。 關鍵字：人類白血球抗原、編碼細胞色素P450、基因型鑑定

The human leukocyte antigen (HLA) is one of the relatively complex regions of the human genome, located on the short arm of chromosome 6 (chr 6p21). The determination of HLA genotypes is crucial for various clinical applications, including organ transplantation, disease diagnosis, and drug allergy reactions. Accurately interpreting the two allele types of each gene from short sequence data obtained from whole genome sequencing is a challenging task. The CYP gene family encodes cytochrome P450 (Cytochrome P450) enzymes, which primarily participate in the metabolism of drugs and endogenous substances in the human body. CYP is one of the most important drug metabolism systems in the human body, with approximately 75% of clinical drugs being metabolized by these enzymes. The accuracy of existing HLA typing tools for Taiwanese individuals still has room for improvement. By incorporating Taiwanese variant information, the quality of the sequences after extraction may increase, thereby affecting the accuracy of HLA allele typing in Taiwanese individuals. This study aims to use HISAT2 and HISAT-genotype tools to filter out 61,424,216 variants extracted from whole genome sequencing data of 1,492 Taiwanese individuals and incorporate them into a graph genome. Based on this graph genome, this study will extend to HLA allele typing. Using HISAT2 tools, indices required for typing 884 individuals' whole genome sequencing data from the Taiwan biobank will be generated. The goal of this study is to evaluate whether the accuracy of HLA genotype typing for seven alleles will improve after incorporating Taiwanese variant information compared to previous studies. The experimental results show that except for a slight increase in HLA-DQA1, the accuracy of the other six alleles has decreased. Additionally, the study will evaluate existing HLA typing tools, T1K and HISAT-genotype, in typing CYP alleles from 44 HPRC individuals' whole genome sequencing data and compare the accuracy with existing CYP typing tools, Aldy and DRAGEN. These HLA typing tools still have room for improvement in CYP allele typing. This study aims to enhance the accuracy of HLA allele typing for the Taiwanese population, contributing to various clinical applications and genetic research, and exploring whether more tools can effectively handle CYP genotype typing issues.

Keywords: human leukocyte antigen, cytochrome P450, genotype identification.