A Novel Partial Sequence Alignment Tool for Finding Large Deletions

Abstract

Finding large deletions in genome sequences has become increasingly more useful in bioinformatics, such as in clinical research and diagnosis. Although there are a number of publically available next generation sequencing mapping and sequence alignment programs, these software packages do not correctly align fragments containing deletions larger than one kb. We present a fast alignment software package, BinaryPartialAlign, that can be used by wet lab scientists to find long structural variations in their experiments. For BinaryPartialAlign, we make use of the Smith-Waterman (SW) algorithm with a binary-search-based approach for alignment with large gaps that we called partial alignment. BinaryPartialAlign implementation is compared with other straight-forward applications of SW. Simulation results on mtDNA fragments demonstrate the effectiveness (runtime and accuracy) of the proposed method.

A comparative analysis of smith-waterman based partial alignment

“…III. THE INCREMENTALPARTIALALIGN ALGORITHM This method takes four input parameters, query sequence, reference sequence, similarity ratio threshold and minimum sequence length [5]. After aligning the query, if the similarity ratio is found to be below the given threshold, considering that this can be due to the presence of a large gap (small gaps would be tolerated by the classical SW anyway), we try splitting the query into a former and a latter part in all possible ways.…”

“…Most NGS software are based on the well-known SmithWaterman (SW) algorithm [2,3,4] are designed to align sequences with small gaps and may not be suitable when large deletions are present [5]. Finding such large deletions (and the related counterpart, finding large insertions), have become increasingly more useful in bioinformatics [5,6,7].…”

“…It has been long known that chromosomal deletions can lead to developmental and malformation disorders [6,8,9] and have a significant role both in the genetics of complex traits such as autism [10] and in genome evolution [11,12]. Figure 1 shows a query pattern that contains a large deletion [5]. The part of the query pattern that comes before the deletion is referred to as the "former part" and the one after the deletion that as the "latter part" of the query.…”

“…In section 3, IncrementalPartialAlign is reviewed, which uses SW in an incremental way, at each iteration advancing the deletion position from one end to the other until convergence. In Section 4, BinaryPartialAlign algorithm is reviewed, which was proposed in [5] to improve the IncrementalPartialAlign method. In Section 5, we present our experimental simulations on mtDNA fragments and we conclude in Section 6.…”

“…Most NGS software are based on the well-known SmithWaterman (SW) algorithm [2,3,4] are designed to align sequences with small gaps and may not be suitable when large deletions are present [5]. Finding such large deletions (and the related counterpart, finding large insertions), have become increasingly more useful in bioinformatics [5,6,7]. It has been long known that chromosomal deletions can lead to developmental and malformation disorders [6,8,9] and have a significant role both in the genetics of complex traits such as autism [10] and in genome evolution [11,12].…”