HW4: Pairwise Alignments

2 minute read



A. Choice of Sequence Type

  • Task 1: Which sequence type - amino acid or nucleotide - is more appropriate to search databases for remotely related sequences? Provide at least three reasons for your decision.

B. Dynamic Programming for Pairwise Alignments

  • Task 2: Create manually (or write an R script for it) one global and one local alignment for the following two protein sequences using the Needleman-Wusch and Smith-Waterman algorithms, respectively:
O15528: PFGFGKRSCMGRRLA
P98187: FIPFSAGPRNCIGQK

Use in each case BLOSUM50 as substitution matrix and 8 as gap extension penalty (no extra penalty for gap opening). Note, here is helper code in R to create the initial matrix programmatically for upload to a spreadsheet program. Alternatively, solve the entire homework by writing an R script. Your answers should contain the following components:

  1. Manually populated dynamic programming matrices
  2. The optimal pairwise alignments created by traceback
  3. The final scores of the alignments

C. Alignments with Different Substitution Matrices

  • Task 1: Load the Biostrings package in R, import the following two cytochrome P450 sequences O15528 and P98187 from NCBI (save as myseq.fasta), and create a global alignment with the pairwiseAlignment function from Biostrings as follows:
library(Biostrings)
myseq <- readAAStringSet("myseq.fasta", "fasta")
(p <- pairwiseAlignment(myseq[[1]], myseq[[2]], type="global", substitutionMatrix="BLOSUM50"))
writePairwiseAlignments(p)

Your answers should address the following items:

  1. Record the scores for the scoring matrices BLOSUM50, BLOSUM62 and BLOSUM80.
  2. How and why do the scores differ for the three scoring matrices?

Homework submission

Accept the homework on Github Classroom, and follow the instructions in the README.md file.

Due date

This homework is due in two weeks on Thu, April 22 at 6:00 PM.

Homework Solutions

See here.

Last modified 2021-04-22: some edits (03f52e457)