---

Polyketide subunit monomer aligner

This repository contains a small program to perform multiple sequence alignment on polyketide backbone structures.

The algorithm is a naive implementation of a progressive multiple sequence alignment, using a guide tree constructed from pairwise alignments (Needleman-Wunsch).

Usage

Creating a multiple sequence alignment

An example input file is shown in ./data/pk_test.fasta:

>seq1
A1A2A1
>seq2
A1A2A1A1
>seq3
B2
>seq4
A2B2A1A1I1
>seq5
A1A2A1
>seq6
CA2A1
>seq7
CB2A1A1I2

A multiple sequence alignment of the sequences displayed above can be performed by running ./scripts/run_multiple_sequence_alignment ./data/pk_test.fasta > ./out/pk_test_msa.fasta from root.

The output file should look like this (- denotes a gap):

>seq2
-A1A2A1A1
>seq1
-A1A2-A1
>seq5
-A1A2-A1
>seq6
-CA2-A1
>seq3
-B2---
>seq4
A2B2A1A1I1
>seq7
CB2A1A1I2

Optional settings

--gap: gap penalty (default: 2)

--end: end gap penalty (default: 2)

Generating new polyketide backbones

A HMM can be constructed from the output MSA file to generate new polyketide backbones with the same conserved motif by running ./scripts/generate_polyketide_backbones.py 1000 > ./out/pk_test_generated_backbones.fasta from root. The number of sequences to generated can be set (in this command: 1000).

Visualizing conserved region in set of polyketide backbones

A stacked barplot logo can be created from MSA to visualize the HMM conserved region by running ./scripts/make_logo_from_msa.py ./out/pk_test_msa.fasta --out ./out/pk_test_hmm_logo.png from root.