Identification of significant conservative and variable regions in homologous protein sequences

A set of aligned homologous protein sequences is divided into 2 groups consisting of m and n sequences. Each group contains sequences from the most related species. Value of the position variability of homologous proteins sequences is defined as a number of failures to coincide in comparison of all possible m.n pairs of amino acid residues (each from a different group) in that position divided by m.n. The variability value plotted vs the sequence position number with a window of 10 positions gives the intergroup variability profile (VP). Area of the figure included between the VP and its mean value line characterizes the overall irregularity of amino acid substitutions along the protein sequences. If the area value S is greater than the average area Sr for 1000 random VPs by more than 2 standard deviation units (sigma), the real VP extrema containing the surplus of area S-(Sr + 2 sigma) are cut off. The cut-off stretches are likely to be significant variable and conservative regions. Intergroup comparisons of protein sequences reveal high overall irregularity of amino acid substitutions and identify variable and conservative regions for all considered families of proteins: phospholipases A2, aspartate aminotransferases, alpha-subunits of Na+,K(+)-ATPase, L- and M-subunits of photosynthetic bacteria photoreaction centre, and human rhodopsins.

A method for the simultaneous alignment of three or more amino acid sequences

We describe an algorithm for the concurrent comparison of three or more amino acid sequences. The basis of the approach is a progressive evaluation of selected segments from each sequence. Only a small subset of all possible segments from each sequence is compared, and a minimum of information is retained for the trace-back of the alignment. As a result, this method has the advantage of being both rapid and minimally consumptive of computer memory when constructing an alignment. This being the case, there are no practical limits on the length of sequences that may be aligned. A computer program for the alignment of three sequences is described, and this method is compared with two three-sequence extensions of the Needleman and Wunsch variety, including a recently published approach. In addition, we have made simultaneous alignments of sets of four and five sequences with this selected-segment method.

Amino acid sequence similarity between rabies virus glycoprotein and snake venom curaremimetic neurotoxins

Evidence was presented earlier that a host-cell receptor for the highly neurotropic rabies virus might be the acetylcholine receptor. The amino acid sequence of the glycoprotein of rabies virus was compared by computer analysis with that of snake venom curaremimetic neurotoxins, potent ligands of the acetylcholine receptor. A statistically significant sequence relation was found between a segment of the rabies glycoprotein and the entire sequence of long neurotoxins. The greatest identity occurs with residues considered most important in neurotoxicity, including those interacting with the acetylcholine binding site of the acetylcholine receptor. Because of the similarity between the glycoprotein and the receptor-binding region of the neurotoxins, this region of the viral glycoprotein may function as a recognition site for the acetylcholine receptor. Direct binding of the rabies virus glycoprotein to the acetylcholine receptor could contribute to the neurotropism of this virus.

Molecular evolution of snake venom toxins

Phylogenetic trees were constructed for 62 venom toxins of snakes of Proteroglyphae suborder using matrix method. The resulting tree from Minimum Spanning Tree-Cluster Analysis technique had the lowest "percent deviation" (8.55). The taxonomic relationship of these toxins agrees very well with zoological opinions. However, the appearance of the tree did not directly provide a plausible evolutionary model for the toxins. A model was derived from nodal ancestral sequence calculations, comparisons between intra- and intergenerical rates of amino acid change, and generally held ideas about protein evolution. According to the model, short neurotoxin is the ancient form of snake venom toxins. The courses of evolution leading to the present intraspecific homologous toxins are explained by gene duplication and allelomorphism.