ROCS: a Reproducibility Index and Confidence Score for Interaction Proteomics Studies

BMC Bioinformatics

Table 5 Overlap and semantic similarity analyses between the ROCS’s lists of Specific Prey Proteins with those obtained by SAINT and ComPASS in all bait experiments (see also Additional file 5 : Table S3)

Bait Experiment	ROCS Unique Prey Proteins(N^B)	ROCS Unique Specific Prey Proteins(S^B)	SAINT Unique Specific Prey Proteins(B)	SAINT Intersection (X)	SAINT Intersection Significance (p-value)	SAINT Semantic Similarity (sim_Res(c₁, c₂))
						BP	MF	CC
STK24	824	112	94	26	1.49146e-05	1.000	1.000	0.853
PPM1B	965	76	86	17	4.62432e-05	0.939	0.936	0.823
NME2	1349	260	210	47	1.25292e-2	1.000	1.000	0.937
CTNNBIP1	1229	69	161	8	1.45195e-1	1.000	0.873	0.937
VHL	5398	43	2	2	5.08701e-05	0.661	0.442	0.307
Bait Experiment	*ROCS* Unique *Prey Proteins*(N^B)	*ROCS* Unique *Specific Prey Proteins*(S^B)	*ComPASS* Unique *Specific Prey Proteins*(B)	*ComPASS* Intersection (X)	*ComPASS* Intersection Significance (p -value)	*ComPASS* Semantic Similarity (*sim*_{Re s}(c₁, c₂))
						BP	MF	CC
STK24	824	112	169	90	4.34990e-57	1.000	1.000	1.000
PPM1B	965	76	159	54	1.07087e-31	1.000	1.000	1.000
NME2	1349	260	242	162	2.50264e-87	1.000	1.000	0.937
CTNNBIP1	1229	69	146	44	2.00480e-27	1.000	1.000	1.000
VHL	5398	43	740	31	2.01445e-20	0.878	0.873	0.823

Using a similar overlap analysis by means of the hypergeometric distribution test and GO semantic similarity as in Table 4, we determined the statistical significance of the overlap between: (Top ) ROCS vs. SAINT, and (Bottom) ROCS vs. ComPASS for all gene ontologies (Biological Process (BP), Molecular Function (MF), and Cellular Component (CC)) and in all bait experiments. SAINT and ComPASS lists are reported for P_SAINT > 0 and D − score > 0 respectively. ROCS lists are reported for C − score C_S > Ĉ_S^cutoffand the marginal inclusion probability thresholds ${\tilde{p}}_{min}^{C}$ and ${\tilde{p}}_{min}^{B}$ as determined in each AP-MS experiment.

ISSN: 1471-2105