Skip to content

Latest commit

 

History

History
111 lines (82 loc) · 2.79 KB

11_KEGG_ANNOTATE_AA.md

File metadata and controls

111 lines (82 loc) · 2.79 KB

KEGG-Annotation

The purpose of this repository is to describe a simple procedure for basic KEGG annotation of AA sequence files

WARNING: THIS TUTORIAL WILL TAKE AT LEAST 10GB OF HARD DRIVE SPACE BECAUSE OF THE KOBAS LIBRARY WARKING: 8GM OF DEDICATED MEMORY ARE NEEDED TO RUN THIS

  • Download the docker bwawrik/bioinformatics:latest
docker pull bwawrik/biorinformatics:latest
  • Make a data directory. If you are working locally on a mac using boot2docker you will need to work in ~/data. Please replace where appropriate
mkdir /data
(* on  your mac: mkdir ~/data)
  • Start the docker and mount /data
docker run -t -i -v /data:/data bwawrik/bioinformatics:latest
(* locally on mac: docker run -t -i -v ~/data:/data bwawrik/bioinformatics:latest)
  • Deploy the KOBAS and Uniprot databses for blast, usearch, and diamond
KOBAS
  • Download KOBAS files and unzip
mkdir /data/KOBAS/sqlite3
mkdir /data/KOBAS/seq_pep
cd /data/KOBAS/sqlite3
wget http://kobas.cbi.pku.edu.cn/download/sqlite3/ko.db.gz
gunzip ko.db.gz
cd /data/KOBAS/seq_pep
wget http://kobas.cbi.pku.edu.cn/download/seq_pep/ko.pep.fasta.gz
gunzip ko.pep.fasta.gz
  • Make a blastable database
makeblastdb -in ko.pep.fasta -dbtype prot
  • Make a Diamond searchable database
diamond makedb --in ko.pep.fasta -d ko.pep

  • Note: the KOBAS fasta file is to big for the free version of Usearch. You will need the 64 bit version to use usearch. This is about $1k/

  • Give everyone permission

sudo chmod 775 *

Download the files needed for matching gIDs to KEGG numbers

wget https://github.com/bwawrik/MBIO5810/raw/master/misc_files/KoAnnotations.gz
wget https://github.com/bwawrik/MBIO5810/raw/master/misc_files/KoGenes.gz
gunzip *.gz

KEGG annotation procedure

  • first, run a diamond search agains the KOBAS KO database
diamond blastp -d /data/DATABASES/KOBAS/seq_pep/ko -q sequence_files/SDB_ONE.faa -o SDB_ONE.faa.dmd -e 1e-10 -k 1
gunzip SDB_ONE.faa.dmd.gz
  • Extract KO numbers and get gene IDs from faa file
cut -f 1,2 SDB_ONE.faa.dmd > SDB_ONE.ORF_G_IDs
grep '>' sequence_files/SDB_ONE.faa | sed 's/>//g' > SDB_ONE.ORF_IDs
cut -f 2 SDB_ONE.ORF_G_IDs > SDB_ONE.G_IDs
  • Make an SQlite database and add the tables
sqlite3 annotate.sqlite

.separator \t
create table ORF_G_IDs (ORF, GID);
.import SDB_ONE.ORF_G_IDs ORF_G_IDs

.separator " "
create table KoGenes (KO, GID);
.import KoGenes KoGenes

.separator \t
create table ORFS (ORF);
.import SDB_ONE.ORF_IDs ORFS

CREATE TABLE IF NOT EXISTS out as
SELECT ORF_G_IDs.ORF, ORF_G_IDs.GID, KoGenes.KO FROM ORF_G_IDs JOIN KoGenes ON ORF_G_IDs.GID = KoGenes.GID;
   
CREATE TABLE IF NOT EXISTS allout as
SELECT * FROM ORFS LEFT JOIN out ON ORFS.ORF = out.ORF

.separator "\t"
.output SDB_ONE.G_ID_KO_ORF_ID
SELECT * FROM allout