08. BLOSUM vs. PAM

As we have seen, there are two different log-odds scoring matrices available. When performing pairwise sequence alignments, you must specify the exact matrix to use - this depends on the suspected degree of identity between the query and its matches.

Let's do a quick summary of these two matrices and see how they differ.

PAM matrices summary

PAM (Accepted Point Mutations) matrices are:

  • Extrapolated using comparisons among closely related proteins based on an evolutionary model.
  • Based on data from alignments of closely related protein families.
  • Because they use the Markov model, the substitution probabilties for highly related proteins can be extrapolated to probabilties of distantly related proteins.

Some limitations of PAM matrices include the somewhat inaccuracy of the assumption model and each position being context dependent.

BLOSUM summary

  • Based on empirical observations of more distantly related protein alignments.
  • Not based on an evolutionary model - just empirically derived.
  • Sequence comparisons cover a broad range of divergences rather than just one subfamily of proteins.
  • Based on observed alignments from conserved regions of multiple sequence alignments (blocks). These blocks are from large protein databases, which give a large sampling site.

Limitations are that you are limited to a subset of conserved domains.

Become a Bioinformatics Whiz!

Bioinformatics Data Skills

Become a Bioinformatics Whiz! Try Bioinformatics

Learn the best practices used by academic and industry professionals. Bioinformatics Data Skills give a great overview to the Linux Command Line, Github, and other essential tools used in the trade. This book bridges the gap between knowing a few programming languages and being able to utilize the tools to analyze large amounts of biological data.

$ Check price
49.9949.99Amazon 4.5 logo(7+ reviews)

More Bioinformatics resources

Learn to be a Pythonista!

Python Playground

Learn to be a Pythonista! Try Python

Python Playground is a collection of fun programming projects that will inspire you to new heights. You'll manipulate images, build simulations, and interact with hardware using Arduino & Raspberry Pi. With each project, you'll get familiarized with leveraging external libraries for specialized tasks, breaking problems into smaller, solvable pieces, and translating algorithms into code.

$ Check price
29.9529.95Amazon 4 logo(14+ reviews)

More Python resources