Optimal reference sequence selection for genome assembly using minimum description length principle

EURASIP Journal on Bioinformatics and Systems Biology

Table 5 Variable number of deletions: the experiment shows the effect of increasing the number of deletions on choice of the reference sequence

Ref. Seq.	SNPs	No. of inversions	No. of insertions	No. of deletions	Code-length using proposed scheme (Kb)
1	0	0	2 / 0	182	1997.28
2	0	0	3 / 0	189	2015.35

The location and length of these deletions was chosen randomly. S_{R 2} has higher number of deletions as opposed to S_{R 1}. The code-length suggests that S_{R 1} is the model of choice as it has a smaller code-length. The experiment show that although no insertions were put in the actual sequence yet still two and three insertions were found for S_{R 1}and S_{R 2}, respectively. This may be due to a large section of reads that could not align to the reference sequence on the edges of these deletions.