Alignment and mapping methodology influence transcript abundance estimation

Table 2 Spearman correlation against ground truth for data simulated using Polyester, incorporating variations into the simulations

Method	PWK	GRCm38.91	StringTie2 assembly
Bowtie2	0.928 ± 0.001	0.919 ± 0.001	0.661 ± 0.001
Bowtie2Strict	0.928 ± 0.001	0.901 ± 0.001	0.653 ± 0.001
Bowtie2rsem	0.932 ± 0.001	0.905 ± 0.001	0.657 ± 0.001
SA	0.931 ± 0.001	0.924 ± 0.001	0.672 ± 0.001
SAF	0.930 ± 0.001	0.922 ± 0.001	0.678 ± 0.001
Quasi	0.930 ± 0.001	0.909 ± 0.001	0.654 ± 0.001
STAR	0.920 ± 0.001	0.912 ± 0.001	0.669 ± 0.001
STARStrict	0.919 ± 0.001	0.887 ± 0.001	0.649 ± 0.001
STARrsem	0.923 ± 0.001	0.890 ± 0.001	0.652 ± 0.001

The first column shows results with reads simulated using the mouse PWK strain and aligned against the PWK strain reference as well. The second column shows results with reads simulated using the mouse PWK strain but aligned against the GRCm38 reference. Hence, the simulated reads have variants and indels with respect to the reference. The third column shows results with reads simulated using the StringTie2 assembled reference, with alignments against the PWK strain reference as input, and then quantified against the GRCm38 reference. Hence, the simulated data contains reads from transcribed, unannotated regions as well

ISSN: 1474-760X