Evolution of protein complexes by duplication of homomeric interactions

Table 1 Data sets investigated in this study

			Pairwise interactions (%)
Dataset	PPIs/Complexes	No. of proteins	HD	PD	F(HD)	F(PD)	F(HI)	F(PI)	Description
Pairwise interactions
Yeast [36]	1,011	753	1.9	13.4					Manual curation of small scale data (does not include yeast two hybrid data)
Yeast-large [37]	15,393	4,741	1.8	6.2					Compilation of small- and large-scale data
Worm [39]	2,422	1,726	1.6	3.3					High-throughput (yeast two-hybrid)
Fly [38]	3,384	2,877	2.9	9.1					High-throughput (yeast two-hybrid)
Complexes
MIPS [36]	216	1,185			32	27			Manual curation
TAP [40]	589	1,474			31	30			High-throughput tagging and mass spectometry
HMS-PCI [41]	741	1,758			33	27			High-throughput tagging and mass spectometry
PQS [29]	2509	3,124					85	11	Three-dimensional structures of protein complexes

PPIs/Complexes are the number of protein-protein interactions and protein complexes (for complexes) in the data sets, respectively. HD, homodimers; PD, dimers of paralogous proteins; F(HD) and F(PD) represent the frequency at which the complexes contain homodimers or dimers of paralogous proteins in any of the two S. cerevisiae protein interaction datasets. These numbers were obtained by computationally superimposing the PINs onto the complex data and are significantly higher than expected by chance at p < 10^-4 in all cases. F(HI) and F(PI) are the frequency of complexes with homomeric or paralogous interactions, respectively.

ISSN: 1474-760X