Please enable JavaScript.

Coggle requires JavaScript to display documents.

Alignments Part 1 - Coggle Diagram

- - - - Multiple alignment far more common in modern bioinformatics yet major multiple alignment algorithms are based on pairwise alignment of all sequences involved
- - - - V few insertions or deletions (useful for v closely related proteins)
    - - A few large insertions (useful for scenarios where entire domain may have been inserted
    - - Many small insertions (useful for more distantly related homologous proteins
- - - - M=11
    - - N= 7
- - - - However, some alignments only conserve common residues (Alanine, proline, serine, threonine)
      - And some conserve less common residues such as W (tryptophan) and Y (tyrosine)
      - Therefore, method for scoring matches btwn AAs which reflects their biological and chemical relationships is needed
    - - Bc cysteine (C) is a relatively rare AA with very particular properties
        
        Whereas serine (S) is relatively common
    - - However, V (valine) to K (lysine)match should be penalised bc the 2 are very dissimilar so unlikely they could do the same job on 2 proteins
      - Thus the fact that diff substitutions occur w/ diff probabilities must be taken into account when attempting to optimise protein sequence alignments
- - - - The PAM matrices developed in 1980s by Margaret Dayhoff
        
        Position Accepted Matrix (PAM) based on global sequence alignment analysis of around 1500 nucleotide mutations in 71 protein families
        
        Proteins chosen were close in evolutionary time with over 85% sequence similarity to reduce the chance of the conclusions being affected by more than one mutation occurring at a particular site
      - The BLOSUM matrices developed by in 1992 by Henikoff
        
        Series of matrices derived from the BLOCKS database using local alignments of around 2000 blocks (conserved motifs) constructed from 504 protein families
      - High numbers in BLOSUM (for closely related seqs) have the opposite significance from higher numbers in PAM (for more distantly related seqs)