Hello, I am looking for options on how to handle the following issue. I am aligning hundreds of amino acid (AA) sequences for HIV-1 Gag. I am interested in the protease cleavage sites in Gag and which positions P12-P1/P1'-P12' are conserved. The problem is, many of the Gag AA sequences have deletions in a cleavage site, typically 2 amino acids. If I remove the 2 amino acid gaps from the aligned Gag sequences, the cleavage site sequences will get shifted to the left or right, either way that will result in a partial misalignment of that cleavage site with the consensus sequence. For example: a cleavage site is from P12-P1/P1'-P12', if P10 and P9 are missing in the Gag sequence and I remove the alignment gaps, P12 and P11 will get shifted to the right into the P10 and P9 positions. HIV-1 produces a lot of defective viruses, so I guess that is what I am seeing. Up until now I have just been deleting the Gag sequences with insertion/deletion, however, that results in the loss of many Gag sequences and I would like to avoid that if possible. How do you handle an issue like this? Thanks in advance.
Hi,
The HIV Databases at LANL (https://hiv.lanl.gov) have tools for looking into this. The genome browser allows you to see each of the Gag cleavage sites, and then you can type in or copy/paste those sites into the QuickAlign/AnalyzeAlign tools to see how variable or conserved each site is. An example of the p24/p2 cleavage site is attached. There are essential no insertions or deletions at that site. KARVLAEAM
Does this method answer your question? Or am I missing something here?
Thanks Brian. Part of the problem is I am looking at cleavage sites spanning P12-P12'. For some Gag cleavage sites like SP1-NC, the sequences have a lot of 2 AA deletions, almost half. See attached screenshot. The SP1-NC cleavage site starts at position 15 with AMSQVTNSAT in the HXB2 consensus sequence at the top. But if you look down the sequences for the N at position 23 you can see lots of seq. have it deleted, same for S at 17 but not as many. I want to present the full cleavage site P12-P12' consensus sequence in a graphical format, but I think the input file has to have an AA at each position, no blanks. Guess I could put in an X in for the deleted AA in order to keep the cleavage sites aligned?, and then just state that in the M&M.
I am not familiar with you naming if the P12/P12' cleavage. HIV-1 (and most of the lentiviruses) have a freakishly high rate of insertion/deletion relative to base substitutions in their evolution, compared to almost all other organisms I have studied. The insetion-deletion rich sites are not random in the genomes. There are specific regions of the genome (Envelope hypervariable loops, some regions of Gag, etc) that seem to be "hot spots" for insertion/deletion.
It may be correct to say an isolate/lineage has a deletion or insertion relative to the majority of other isolates/strains in most of these sites, but very often the it is essentially impossible to say which way the alignment should be done to push the gaps left or right. The "right way" for a phylogenetic/evolution study may not be the "right way" for studying the 3D structure of the protein, or peptides targeted by T-cells.
For the envelope hypervariable loops, we created tool to study the properties of the loop sequences (length of each loop, number of potential N-linked glycosylation sites per loop, etc.) but we have not done that for other regions of the genome. The attached GagP2PROT.fasta file here is aligned a bit differently than your example. I deleted the SIV-Chimpanzee, SIV-Gorilla, and the HIV-1 N, O and P samples, to leave just the HIV-1 M group.
The attached "LOGO" image is from our QuickAling/AnalyzeAlign tool for this alignment.
- Badges
- Science topic
- Similar topics
- Geoscience
- Asia
- India