A simple and economical method for improving whole genome alignment

Huijun Mai,Tak-Wah Lam,Hing-Fung Ting

doi:10.1186/s12864-017-3734-2

Abstract

BackgroundThe recent advancement of whole genome alignment software has made it possible to align two genomes very efficiently and with only a small sacrifice in sensitivity. Yet it becomes very slow if the extra sensitivity is needed. This paper proposes a simple but effective method to improve the sensitivity of existing whole-genome alignment software without paying much extra running time.Results and conclusionsWe have applied our method to a popular whole genome alignment tool LAST, and we called the resulting tool LASTM. Experimental results showed that LASTM could find more high quality alignments with a little extra running time. For example, when comparing human and mouse genomes, to produce the similar number of alignments with similar average length and similarity, LASTM was about three times faster than LAST. We conclude that our method can be used to improve the sensitivity, and the extra time it takes is small, and thus it is worthwhile to be implemented in existing tools.

Highlights

The recent advancement of whole genome alignment software has made it possible to align two genomes very efficiently and with only a small sacrifice in sensitivity
This paper proposes a simple but effective method to improve the sensitivity of existing whole-genome alignment software without paying much extra running time
We have applied it to improve LAST, and our results show that it can increase the number of LAST’s reported alignments to the one that LAST gets with threshold 0.85D, but using time similar to that LAST uses with threshold D

Summary

Results and conclusions

We have applied our method to a popular whole genome alignment tool LAST, and we called the resulting tool LASTM. Experimental results showed that LASTM could find more high quality alignments with a little extra running time. When comparing human and mouse genomes, to produce the similar number of alignments with similar average length and similarity, LASTM was about three times faster than LAST. We conclude that our method can be used to improve the sensitivity, and the extra time it takes is small, and it is worthwhile to be implemented in existing tools

Background

Methods

Results and discussion

Conclusions