BMC Bioinformatics

Table 2 Effect of different choices for the possible pipelines in TCC

From: Evaluation of methods for differential expression analysis on multi-group RNA-seq count data

PG1	33 %	50 %	50 %	60 %	60 %	70 %	80 %
PG2	33 %	30 %	40 %	20 %	30 %	20 %	10 %
PG3	33 %	20 %	10 %	20 %	10 %	10 %	10 %
(a) P_DEG = 5 %
*EEE-E*	91.57	91.50	91.50	91.43	91.42	91.45	91.46
DED-E	91.57	91.50	91.50	91.43	91.42	91.46	91.47
EDE-E	91.57	91.50	91.50	91.43	91.42	91.45	91.46
DDD-E	91.57	91.50	91.50	91.43	91.42	91.45	91.46
EEE-D	90.70	90.62	90.64	90.54	90.55	90.58	90.62
DED-D	90.71	90.62	90.64	90.54	90.55	90.59	90.62
EDE-D	90.70	90.62	90.64	90.54	90.55	90.58	90.62
*DDD-D*	90.70	90.62	90.64	90.54	90.55	90.59	90.62
*E-E* (edgeR)	91.58	91.48	91.47	91.38	91.37	91.38	91.34
D-E	91.58	91.48	91.46	91.38	91.36	91.36	91.32
E-D	90.70	90.61	90.61	90.50	90.50	90.51	90.50
*D-D* (DESeq)	90.71	90.60	90.60	90.50	90.49	90.50	90.48
(b) P_DEG = 25 %
*EEE-E*	91.47	91.46	91.45	91.45	91.43	91.42	91.37
DED-E	91.47	91.46	91.47	91.47	91.45	91.45	91.43
EDE-E	91.47	91.43	91.41	91.40	91.36	91.30	91.19
DDD-E	91.47	91.44	91.43	91.42	91.39	91.36	91.29
EEE-D	90.77	90.74	90.74	90.73	90.71	90.71	90.65
DED-D	90.77	90.74	90.76	90.75	90.73	90.74	90.71
EDE-D	90.77	90.71	90.70	90.68	90.64	90.60	90.47
*DDD-D*	90.77	90.73	90.72	90.70	90.68	90.65	90.57
*E-E* (edgeR)	91.47	91.30	91.18	91.06	90.98	90.62	89.97
D-E	91.48	91.25	91.08	90.96	90.86	90.44	89.75
E-D	90.77	90.59	90.48	90.35	90.26	89.92	89.25
*D-D* (DESeq)	90.77	90.54	90.37	90.25	90.15	89.73	89.04

Legends are basically the same as in Table 1. Results of a total of 12 pipelines are shown. The AUC values for four pipelines (EEE-E, DDD-D, E-E, and D-D) in bold are also shown in Table 1. The DED-E pipeline outperforms the others overall

Back to article page

ISSN: 1471-2105

Contact us

General enquiries: journalsubmissions@springernature.com