Personal communication to FlyBase, 24 Jan 2007 Thomas Sheldon Division of Mathematical Biology National Institute for Medical Research The Ridgeway, Mill Hill, London NW7 1AA tsheldon@XXXX +44 (0)20 8816 2300 (phone) +44 (0)20 8816 2460 (fax) The protein product for Dm gene CG12677 was replaced by a new sequence in 2006, yet the original already fitted the VRP protein family I am investigating in terms of its canonical sequence identity, intron phasing and signal peptide. The unusual response pattern of its gene is also consistent with the expression analysis for the other Dm genes. The C-terminus of D. melanogaster protein CG6301 is identified as a family member. It conforms to the canonical sequence, despite the existing record showing a single protein of 463 amino acids. Rather than an isolated case of a family member embedded as a domain in a larger protein, closer inspection of the deposited cDNA fragments for the record show that the N-terminus section has a poly-A tail, indicating that the predicted protein should in fact be two. In support of this claim is the Pfam entry for DUF745, to which a portion of the N-terminus of CG6301 (residues 94-302) is correctly predicted to belong, but where the C-terminus sequence fragment (containing the first of the cysteine residues) is incorrect. This new protein is 192 residues long, and consistent with the rest of the family it has a short signal peptide at its N-terminus showing secretedness. Dm gene CG14132 is currently annotated without a signal peptide sequence. A 5' upstream coding exon has now been found and the sequence will be updated, bringing it into line with the predicted sequence RE17110p in GenBank. Dm protein BK003543 is predicted in GenBank but the corresponding cDNA record in FlyBase may be artefactual given its length. In support of its existence, however, it receives a Blast e-value of 3e-5 against protein CG6301, and displays the canonical pattern common to all VRP proteins.
