ndreey
commented
1 month ago Creating an anvio contigs database
An anvio contigs database will keep all the information related to your contigs: positions of open reading frames, k-mer frequencies for each contigs, where splits start and end, functional and taxonomic annotation of genes, etc. The contigs database is an essential component of everything related to anvi’o metagenomic workflow.
The anvi-gen-contigs-database command will:
- Compute k-mer frequencies for each contig. The default is 4 but can be changed with
--kmer-size. - Soft-split contigs longer than 20kbp into smaller ones. When the gene calling step is not skipped, the process of splitting contigs will consider where genes are and avoid cutting genes in the middle.
- Identify ORF using
Prodigal.
Furthermore we will look for SCG using hidden makarov models, get functional annotation using NCBI-COGS as well as getting taxonomy for the contigs. I used this script: contig-db.sh
#!/bin/bash
#SBATCH --job-name anvio-contigdb
#SBATCH --array=1-2
#SBATCH -A naiss2024-22-580
#SBATCH -p core -n 12
#SBATCH -t 01:30:00
#SBATCH --output=slurm-logs/anvio/contig-db/SLURM-%j-contigdb-%a.out
#SBATCH --error=slurm-logs/anvio/contig-db/SLURM-%j-contigdb-%a.err
#SBATCH --mail-user=andbou95@gmail.com
#SBATCH --mail-type=ALL
# Start time and date
echo "$(date) [Start]"
# Activate the environment
mamba activate anvio-8
# SLURM array jobid
JOBID=${SLURM_ARRAY_TASK_ID}
# Assign POP based on JOBID
if [ "$JOBID" -eq 1 ]; then
POP="CHST"
elif [ "$JOBID" -eq 2 ]; then
POP="COGE"
else
echo "Invalid JOBID: $JOBID"
exit 1
fi
# Paths and variables
FIX_DIR=00-FIXED-ASSEMBLY
DB_DIR=02-CONTIG-DB
CDB=${DB_DIR}/${POP}.db
NUM_THREADS=12
COG_DIR=/crex/proj/snic2020-6-222/Projects/Tconura/working/Andre/databases/ncbi-cogs
SCG_DIR=/crex/proj/snic2020-6-222/Projects/Tconura/working/Andre/databases/scg/
# Move to the anvio working directory
cd 07-ANVIO
# Generate folder for contig database
if [ ! -d "${DB_DIR}" ]; then
mkdir -p ${DB_DIR}
fi
# Generate the anvio contig database
anvi-gen-contigs-database \
--contigs-fasta ${FIX_DIR}/${POP}-contigs.fa \
--project-name Tconura_${POP} \
--num-threads ${NUM_THREADS} \
--output-db-path ${DB_DIR}/${POP}.db \
--description ../doc/short-project-desc.txt
# Run hidden makarov models.
anvi-run-hmms \
--contigs-db ${CDB} \
--also-scan-trnas \
--num-threads ${NUM_THREADS}
# Annotate the genes with functions from COG database
anvi-run-ncbi-cogs \
--contigs-db ${CDB} \
--cog-version COG20 \
--cog-data-dir ${COG_DIR} \
--num-threads ${NUM_THREADS}
# Classify the taxa for the database
anvi-run-scg-taxonomy \
--contigs-db ${CDB} \
--scgs-taxonomy-data-dir ${SCG_DIR} \
--num-threads ${NUM_THREADS}
# Estimate the taxonomy
anvi-estimate-scg-taxonomy \
--contigs-db ${CDB} \
--metagenome-mode \
--num-threads ${NUM_THREADS}
# End time and date
echo "$(date) ${POP} [End]"
Following the installation of Anvio (#23) we can now run it by activating the mamba environment
anvio-8.Lets create a new directory for the anvio run that we will use as the working dir.
Re-formatting the assembly fasta and mapping
To avoid complicated fasta headers we will use
anvi-script-reformat-fastato change the headers to a more standardized format that anvio can handle. This must be done before mapping the reads to the assembly. The script will then map and sort the.bamfiles for the new reformatted assemblies.CHST results
COGE results