Table 1.

Biological datasets used for RMRCM performance asssessment

Dataset	Nsetsa	Nprotb	Ncolb	Structurec	Reference
PFAM	1256	200/3975	10/521	Severald	[31]
CASP9	28	5/501	21/227	Severale	www.predictioncenter.org
MADS	12	34/339	78/218	1n6j	[36]
Response regulators	1	1433	186	1xhe	[16]
CDD	36	125/1922	34/411	Severalf	[8]
SK-RR	1	4934	184	2c2a, 1pey, 1f51	[16]
PDZ-peptide	1	2385	162	1n7f	[33]

^a N_sets, number of separate multiple sequence alignments. Four datasets consists of several multiple sequence alignments, each of which is analyzed separately. For these, the number of proteins and the number of columns mentioned are the minimum and maximum found in these sets.

^b N_prot, number of proteins; N_col, number of columns in the multiple sequence alignment.

^c PDB identifier of structure used to compare predicted residue contacts.

^d Obtained via PFAM.

^e Obtained via www.predictioncenter.org.

^f See ref. [8].