noreaster/bioinf527proj

Recalculation of BLOSUM using modern sequence data

Recalculation of BLOSUM using modern sequence data

Files in this Repository:

• 1234.txt results from homologsearch.py for seed 1234
• 5678.txt results from homologsearch.py for seed 5678
• 9001.txt results from homologsearch.py for seed 9001
• astral-scopedom-seqres-gd-sel-gs-bib-49-2.06.fa sequences from the Astral database for testing
• blosum62.csv results for blosum62
• blosum62.mat the BLOSUM62 matrix used in NWalign
• homologsearch.py runs NW comparisons and generates confusion matrices for each substitution matrix
• iij2mat.py converts files output by blosum.c into format readable by NWalign
• makefile makefile for NWalign.cpp
• new99.csv results for new99
• new99.mat our new generated matrix used in NWalign
• NWalign.cpp Needleman-Wunsch alignment implemented in C++ for efficiency
• NWalign.h header file for NWalign.cpp
• NWalign.py original NW implementation from which NWalign.cpp was made
• randblos.csv results for randblos
• randblos.mat randomly generated substitution matrix based off of BLOSUM62
• randblos.py randomly generates a subsitution matrix given value frequencies present in BLOSUM62
• README.md this file
• seeds.txt the three seeds used to run triplicate
• tests.R R script that generates mean accuracy, precision, and recall and performs t-tests of all these parameters between the three matrices