Skip to main content

Table 1 Performance on simulated E. coli dataset

From: Kohdista: an efficient method to index and query possible Rmap alignments

Method

Time (s)

Memory (MB)

Align-ments

Recall

Precision

Kohdista

20

19.0

907

702/4305 (16%)

702/771 (91%)

Kohdista (lax)

373

18.3

8545

3925/4305 (91%)

3925/8545 (46%)

Valouev et al.

148

4.0

742

699/4305 (16%)

699/742 (94%)

MalignerDP

47

6.0

1959

1296/4305 (30%)

1296/1959 (66%)

OMBlast

116

2078

1008

806/4305 (19%)

806/1008 (80%)

RefAligner

31

81.2

992

958/4305 (22%)

948/992 (97%)

MalignerIX

4

6.0

0

0/4305 (0%)

0/0 (N/A)

OPTIMA

455

10756.5

0

0/4305 (0%)

0/0 (N/A)

  1. Kohdista (lax) demonstrates that our indexing and search method is capable of finding the majority of ground truth alignments when the search is pruned to the more relaxed thresholds for chi-squared-cdf-thresh and binom-cdf-thresh, i.e., chi-squared-cdf-thresh = 0.02 and binom-cdf-thresh = 0.5