virus_discoveries.txt

Question 1
I used blastp, protein version of the Basic Local Alignment Search Tool.
It aligns the sequence to sequences of known proteins using the database RefSeq non-redundant proteins.
seq_1 was identified as a caspid protein from Adeno-associated virus 4 with 100% coverage.
seq_2 was identified as a caspid prtein VP1 from Adeno-associated virus with 100% coverage.

Question 2
Using Python3, I coded createmultifasta.py. It takes as argument a file, which contains the file names of
the sequences and the results from blastp, and the output name.
python3 createmultifasta.py -i seqNres.txt -o outputname.fasta

Question 3
I used Clustal Omega from the EBI for the multi-sequence alignment as it uses Hidden Markov models for the alignment.
HMMs have been proven as one of the best method to align multiple sequences together. In the past, I have used
HHblits which uses also HMM. However, it does not have a web interface. Clustal Omega is good in that way to be
able to run a quick alignment. For several sequences, I would probably take another approach, coding wise.

Question 4
A simple approach of recording the similarities and differences is counting the percentage of identities, gaps
and mismatch. This approach can be improved by looking at percentage of identity of the active site, as it is
the most important part of the protein. From the sequence files and their NCBI page, we do not have the information
of the active sit localisation. Therefore, I can only count the similarities and differencies. 

Question 5
Alignment similarities and differences :
Percentage of conserved residues :  62.38 %
Percentage of strongly conserved residue properties :  12.85 %
Percentage of weakly conserved residue properties :  8.3 %
Percentage of residues not conserved :  16.47 %
Percentage of overall conservation :  83.53 %
seq_1 and seq_2 have an overall residue conservation of 83.53%, which is very high. We can conclude that the two proteins 
are homologous in the two viruses.

Question 6
As the two proteins are caspid proteins, they probably have the same function and interactome. It would be 
interesting to look at the difference of structure using pdb files and pdb viewer. The structures don't have to be
exactly similar to be homologous proteins. The structure is important as key places such as interaction sites. If 
the structure is different at one of these sites, that would mean they don't interact with the same molecules and
probably don't have the same function in the two viruses.