Selection of Optimal Bioinformatic Tools and Proper Reference for Reducing the Alignment Error in Targeted Sequencing Data

Abstract

Background: Careful design in the primary steps of a next-generation sequencing study is critical for obtaining successful results in downstream analysis. Methods: In this study, a framework is proposed to evaluate and improve the sequence mapping in targeted regions of the reference genome. In this regard, simulated short reads were produced from the coding regions of the human genome and mapped to a Customized Target-Based Reference (CTBR) by the alignment tools that have been introduced recently. The short reads produced by different sequencing technologies aligned to the standard genome and also CTBR with and without well-defined mutation types where the amount of unmapped and misaligned reads and runtime was measured for comparison. Results: The results showed that the mapping accuracy of the reads generated from Illumina Hiseq2500 using Stampy as the alignment tool whenever the CTBR was used as reference was significantly better than other evaluated pipelines. Using CTBR for alignment significantly decreased the mapping error in comparison to other expanded or more limited references. While intentional mutations were imported in the reads, Stampy showed the minimum error of 1.67 using CTBR. However, the lowest error obtained by stampy too using whole genuine and one chromosome as references was 3.78 and 20, respectively. Maximum and minimum misalignment errors were observed on chromosome Y and 20, respectively. Conclusion: Therefore using the proposed framework in a clinical targeted sequencing study may lead to predict the error and improve the performance of variant calling regarding the genomic regions targeted in a clinical study.