Novel approach for accurate detection of contaminating human mitochondrial DNA in next-generation sequencing data

Abstract

DNA molecules are degraded after the death of an organism. However, the degree and rate of DNA degradation enormously vary depending on environmental conditions, such as temperature or humidity, which greatly affect DNA preservation. Most samples excavated in warm, humid, or dry areas are often poorly preserved samples with from <0.1%-1% endogenous DNA. In these degraded samples, the contamination by exogenous DNA remains a potential challenge, no matter how much effort is made to prevent it. For an accurate DNA sequence analysis, quality control must be thoroughly performed, using the mitochondrial DNA as an indicator of exogenous DNA contamination. Here, we propose a practical approach for detecting exogenous human mitochondrial macro haplogroups, and discuss the effectiveness of this approach using simulated data. Our approach is based on the Bayes classification, which is a supervised machine learning algorithm, and it can detect a contaminating macro haplogroup in high-throughput sequencing data. This approach can help validate the quality of high-throughput sequencing data from possibly contaminated or degraded human samples.