Mismatch repair contributes to the overall fidelity of DNA replication [1]. It involves the correction of mismatched base pairs that have been missed by the proofreading element of the DNA polymerase complex. The sequence of some proteins involved in mismatch repair in different organisms have been found to be evolutionary related [2,3]. One of these families is called mutS [4], it consists of:

• Prokaroytic protein mutS protein (also called hexA in Streptococcus pneumoniae). Muts is thought to carry out the mismatch recognition step of DNA repair.
• Eukaryotic MSH1, which is involved in mitochondrial DNA repair.
• Eukaryotic MSH2, which is involved in nuclear postreplication mismatch repair. MSH2 heterodimerizes with MSH6. In man, MSH2 is involved in a form of familial hereditary nonpolyposis colon cancer (HNPCC).
• Eukaryotic MSH3, which is probably involved in the repair of large loops.
• Eukaryotic MSH4, which is involved in meiotic recombination.
• Eukaryotic MSH5, which is involved in meiotic recombination.
• Eukaryotic MSH6 (also known as G/T mismatch binding protein), a DNA-repair protein that binds to G/T mismatches through heterodimerization with MSH2.
• Prokaryotic protein mutS2 whose function is not yet known.
• A coral (Sarcophyton glaucum) mitochondrial encoded mutS-like protein.

As a signature pattern for this class of mismatch repair proteins we selected a region rich in glycine and negatively charged residues. This region is found in the C-terminal section of these proteins; about 80 residues to the C-terminal of an ATP-binding site motif 'A' (P-loop) (see <PDOC00017>).