chenyangkang / Fasta2Codeml

A codeml (PAML package) wrapper to make life easier. Dummy input unaligned multi-species fasta file (a single gene), and output codeml result.

Fasta2Codeml

A codeml (PAML package) wrapper to make life easier. Dummy input unaligned multi-species fasta file (a single gene), and output codeml result.

Prerequisites

1. Codeml (PAML version 4.10.6)

2. MACSE (.jar form)

3. MUSCLE

4. RAXML

5. biopython (v1.81, python package)

6. newick (v1.9.0, python package)

must be installed beforehand

Installation

Simply add ./script to your environment

Input data

1. A single gene fasta sequence file (multi-species, not aligned).
2. A text file which indicate the foreground species. One species each line.

Example

cd to example/test_space/

Change the absolute path in the command lines below to to your path.

For single fasta file mode

type:

Fasta2Codeml.py \
--out_dir /beegfs/store4/chenyangkang/DEV/Fasta2Codeml/example/test_space_single_gene \
--project_name Simple_test \
--foreground_file /beegfs/store4/chenyangkang/DEV/Fasta2Codeml/example/foreground.txt \
--fasta /beegfs/store4/chenyangkang/DEV/Fasta2Codeml/example/single_gene/CLOCK.fasta \
--muscle /beegfs/store4/chenyangkang/software/ParaAT2.0/muscle \
--macse /beegfs/store4/chenyangkang/software/macse_v2.07.jar \
--raxml /beegfs/store4/chenyangkang/software/standard-RAxML/raxml \
--codeml /beegfs/store4/chenyangkang/miniconda3/bin/codeml \
--boostrap 10 \
--codon_frac 0.5 \
--sp_frac 0.5

For multi-file model (multi-cds file mode):

Fasta2Codeml.py \
--out_dir /beegfs/store4/chenyangkang/DEV/Fasta2Codeml/example/test_space_multi_cds \
--project_name Simple_multi_test \
--foreground_file /beegfs/store4/chenyangkang/DEV/Fasta2Codeml/example/foreground.txt \
--multi_file \
--multi_file_list cds_list.txt \
--muscle /beegfs/store4/chenyangkang/software/ParaAT2.0/muscle \
--macse /beegfs/store4/chenyangkang/software/macse_v2.07.jar \
--raxml /beegfs/store4/chenyangkang/software/standard-RAxML/raxml \
--codeml /beegfs/store4/chenyangkang/miniconda3/bin/codeml \
--boostrap 10 \
--codon_frac 0.5 \
--sp_frac 0.5

Workflow underneath

1. Remove species that contain only "N"s.
2. Run muscle alignment with 5 iterations.
3. Refine alignment using MACSE.
4. Replace frameshift(!) and stop codon with NNN using MACSE.
5. Concatenate files (if in multi-file mode).
6. Remove codon columns with more than 50% species missed, and remove species with more than 50% codons as "NNN" or "---".
7. Build tree with raxml -f a -x 42 -p 42 -m GTRGAMMA .
8. Co-filter fasta file and tree file. Trim and annotate tree with the foreground information provided. Output alignment as phylip format.
9. Generate codeml configuration files for both branch-site null model (omega=1) and alternative model.
10. Run both codeml model.
11. Generate p values and other statistics using scipy.

Disclaimer

This package is not liable for any ouput results and scientific/commercial interpretation (see the MIT license).