Four groups of AMP are known in mussels, defensins, mytilins myticins and mytimycins. The cationic and amphipatic structure of the mature peptides is stabilized by 4 intrachain disulphide bonds according to a unifying tridimensional motif. Mytibase includes the full length precursor sequences of all the mussel AMP with some new var iants, they are reported as mature peptide sequences in Figure 1. Myticins are subdivided in Inhibitors,Modulators,Libraries A, B and the polymorphic type C. Searching tBLASTn similarities to prototype sequences, we identified in Mytibase many precursors of myticin C, myticin A and myticin B. Robust non synonymous SNPs analysis allowed us to split the sequence cluster of myticin A into 5 subgroups named A, A2, A3, A4 and A5, confirmed by 23, 38, 2, 21 and 4 sequence traces of high quality, respectively.

Mytilin precursors are more heterogeneous in length ranging between 97 and 105 residues, and can be easily differentiated from the myticin precursors due to a dif ferent cysteine pattern. Similarly, we identified mytilin Inhibitors,Modulators,Libraries A, mytilin B, myti lin C, mytilin D. We could also extend the sequence of Mytilin Dacomitinib G1 and we propose MGC00659 as Mytilin F, namely a new myti lin component. The defensin precursors identified in Mytibase are MGD1, MGD2b and three new sequences proposed as MGD3, MGD4, and MGD5. Due to the presence of a stop codon just after the 8th conserved cystein, defensins MGD3 and MGD4 are shorter than the others whereas MGD5 is the longest with 97 aminoacid residues. Only one Mytibase EST corre sponds to the mytimycin described in M.

edulis and four other sequences Inhibitors,Modulators,Libraries grouped from 4, 4, 4 and 3 ESTs may be regarded as new mytimycin variants. Curiously, two of these ESTs display a long insertion in the 5 UTR and a signal peptide with maximal cleavage prob abilities between positions 18 24 from ATG. cDNAs normalization was essential to reveal Inhibitors,Modulators,Libraries the rare mytimycin ESTs whereas the other more abundant AMP sequences can be easily and mainly attributed to hemocyte libraries prepared from immunostimulated Italian and Spanish mussels, without evidence of preferential geographical distribution. All mussel AMP and one hydramacin like transcript have been included in the Immunochip. Transcripts containing C1q and Tumour Necrosis Factor like domains The overlapping C1q and TNF like domains have probably evolved by diver gence from an ancient recognition molecule whose diversification could have started with urochordates and cephalocordates. The large family of proteins with a C1q domain support many biological processes, from complement activation, modulatory immune func tions, apoptotic cell clearance to coagulation, embryonic development and tissue homeostasis.