Table 1. Statistics of PDB-PMID-Residue relationships in CSA.

PDB = Protein Data Bank. CSA = Catalytic Site Atlas. CSA-Lit = the subset of CSA annotations marked as based on literature. PMID = PubMed ID. A verified text residue is a residue that has been identified through text mining, and mapped to a physical residue in the corresponding PDB protein sequence. “Site” refers to a particular numbered location in a protein sequence.

Source	Set	Residues	PDB	PMIDs	(PMID,Site)
1. PDB	PDB residues, with abstract	17904740	30816	17595	4797110
2. PDB	PDB residues with verified text residues (abstracts)	44701	9923	5236	14127
3. PDB	PDB residues with verified text residues (full text)			7309	107153
4. CSA	PDB residues in CSA	112031	17524
5. CSA	PDB residues in CSA, with abstract	94327	14673	7587	29447
6. CSA	Verified text residues; match to CSA (abstracts)	9059	3163	1630	2708
7. CSA-Lit	PDB residues in CSA-Lit	6372	942
8. CSA-Lit	PDB residues in CSA-Lit, with abstract	5586	831	823	2799
9. CSA-Lit	PDB residues in CSA-Lit, with abstract with at least one verified text residue	2116	343	341	1139
10. CSA-Lit	Verified text residues; match to CSA-Lit (abstracts)	878	259	259	476
11. CSA-Lit	Verified text residues; match to CSA-Lit (full text)			312	805
12. CSA-Lit	Verified text residues; match to CSA-Lit (full text + abstract)			444	1052