Improving the sensitivity of long read overlap detection using grouped short k-mer matches

BMC Genomics

Table 3 Overlap detection on the human metagenomic dataset

	GroupK	Minimap	Minimap2	DALIGNER	GraphMap	MHAP
Total:
Best F1 score	0.9163	0.8306	0.8776	0.6911	0.7188	0.7812
Sensitivity	0.8954	0.7802	0.8352	0.6012	0.5721	0.6803
Precision	0.9381	0.8880	0.9245	0.8027	0.9666	0.9174
C. aurimucosum:
Best F1 score	0.9512	0.8072	0.9117	0.7432	0.7397	0.8545
Sensitivity	0.9228	0.6858	0.8467	0.6266	0.5892	0.8045
Precision	0.9814	0.9806	0.9874	0.9131	0.9937	0.9111
C. tuberculostearicum:
Best F1 score	0.9454	0.8688	0.9050	0.8315	0.7276	0.7961
Sensitivity	0.9105	0.7958	0.8346	0.7150	0.5727	0.7355
Precision	0.9830	0.9567	0.9884	0.9934	0.9977	0.8675
S. hominis:
Best F1 score	0.9163	0.7651	0.8024	0.8754	0.6343	0.6861
Sensitivity	0.8733	0.6887	0.6813	0.7967	0.4658	0.6348
Precision	0.9245	0.8606	0.9759	0.9713	0.9938	0.7464

The performance of overlap detection using GroupK, Minimap, Minimap2, DALIGNER, MHAP, and GraphMap on the mock metagenomic dataset. Here we only report the experimental results with the highest F1 score for each tool

ISSN: 1471-2164