I am looking for a residue in a peptide for substitution using the information from BLOSUM matrix. I do not like to go for alignment of peptides with homologs.
Any help will be appreciated.
Sir select the one with highest score
for example consider the blosum62 matrix (http://zhanglab.ccmb.med.umich.edu/NW-align/BLOSUM62.txt)
If you want to substitute Leucine with some other residue then your first option should be either Isoleucine or methionine (as both have score of +2 with leucine) and second option should be valine (having score of +1 with leucine).
Hope it helps :)
Thanks Sandy,
This matrix will help us once we have finalized a residue (like Leucine in your example). I am interested in that how we can select a residue out of 15mer peptide?
In other words, substitution at which position in a peptide would effect least to its functional activity.
Suppose if sequence is "SANDEEP" then alignment of SANDEEP with SANDEEP will score 37 as per blosum62 matrix scoring function.
Now any mutation will lead to lowering the total alignment score but select that mutation which will least lower the score should be done. In case of
SANDEEP (if considering only 20 mutation types and not considering columns B Z and X in the blosum62 matrix) the mutation effecting
least in principle should be either S mutated to A/N/T (total score will be 34), or A mutated to S (total score 34) or E mutated to D/Q (total
score 34). Any other mutation will have score less than 34.
Thanks Sandy.. The only issue to follow this idea is that it will not tell the position of substitution, when we have same residue at two positions, like E (at 5th and 6th position) in the example of SANDEEP. Let me extend the question further, by providing an example of a publication where similar analysis was done.
Article Optimization algorithms for functional deimmunization of the...
The paper you shared have described the same thing as mentioned above. (See section BLOSUM) and they have mentioned that "we obtain a reasonably conservative set of acceptable residues by only taking those with score differences of at most 4". At least this approach gives you a set of mutated protein analogs which can be further screened using prediction tools.
Thanks Sandy for your effort to dig into answer. In this way, we will be having a set up of peptides and I am looking for a way to end up with single peptide.
- Badges
- Science topic
- Similar topics
- Biological Science
- Biochemistry
- Biomolecules
- Peptides