Scholarlink

Abstract <p>Many modern human populations contain archaic Neanderthal segments through the archaic introgression event, which occurred 45,000–55,000 years ago. There are a few computational methods that could infer such tracts. The results of their inference is used in the downstream analysis to better understand the impact of the archaic component on the genetic diversity of modern humans, its genetic costs (in particular in term of health), to refine our joint population history, etc. In this paper we present a methodology for numerically evaluating the sources of errors which might affect downstream data analysis. Using simulations, we compare the performance of our recently developed method <Emphasis FontCategory="NonProportional">DAIseg</Emphasis> and the previous state-of-the-art method <Emphasis FontCategory="NonProportional">hmmix</Emphasis>. We show that <Emphasis FontCategory="NonProportional">DAIseg</Emphasis> accurately reconstructs archaic tracts even with relatively low quality archaic genomes. We also confirm that most of the errors come from the short tracts which are very noisy to infer. Overall, our results can be used both as a guideline, and as a methodology for validating the use of the methods for archaic tract inference for particular downstream tasks.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

How Much Could We Trust the Inferred Neanderthal Segments?

  • 期刊论文

  • 发表日期：

    2025年7月1日

查看全文
  • A. V. Ilina,
  • L. Planche,
  • V. L. Shchur

摘要

Abstract

Many modern human populations contain archaic Neanderthal segments through the archaic introgression event, which occurred 45,000–55,000 years ago. There are a few computational methods that could infer such tracts. The results of their inference is used in the downstream analysis to better understand the impact of the archaic component on the genetic diversity of modern humans, its genetic costs (in particular in term of health), to refine our joint population history, etc. In this paper we present a methodology for numerically evaluating the sources of errors which might affect downstream data analysis. Using simulations, we compare the performance of our recently developed method DAIseg and the previous state-of-the-art method hmmix. We show that DAIseg accurately reconstructs archaic tracts even with relatively low quality archaic genomes. We also confirm that most of the errors come from the short tracts which are very noisy to infer. Overall, our results can be used both as a guideline, and as a methodology for validating the use of the methods for archaic tract inference for particular downstream tasks.