[PSC] BLAST example #bridges2

blastn-example.sh

#!/bin/bash
#SBATCH --job-name=blast_demo
#SBATCH --output=blast_%j.out
#SBATCH --error=blast_%j.err
#SBATCH --time=02:00:00
#SBATCH -n 8
#SBATCH --mem=16000M
#SBATCH -p RM-shared

###############################################################################
# blast_demo.sh
#
# Copyright (c) 2026 icaoberg at the Pittsburgh Computing Center for Bridges-2
#
# Description:
#   Example BLASTN search against the human genome (GRCh38) using a multi-volume
#   BLAST v5 database. Generates a guaranteed matching query sequence from the
#   database and runs blastn.
#
# Date: 2026-02-11
###############################################################################

echo "Loading BLAST module..."
module load BLAST

echo "Extracting human genome BLAST database..."
tar -xvzf /ocean/datasets/community/genomics/blast/latest/human_genome.00.tar.gz
tar -xvzf /ocean/datasets/community/genomics/blast/latest/human_genome.01.tar.gz
DATABASE=GCF_000001405.39_top_level

echo "Generating guaranteed matching query sequence..."
# Select a valid accession from the database to guarantee a match
SEQID=$(blastdbcmd -db $DATABASE -entry all -outfmt "%a" | head -n 1)
echo "Using sequence: $SEQID"
blastdbcmd -db $DATABASE -entry "$SEQID" -range 1-200 > query.fa

echo "Running blastn..."
blastn \
  -query query.fa \
  -db $DATABASE \
  -out blast_results.txt \
  -evalue 1e-5 \
  -outfmt 6 \
  -num_threads 8

echo "Top BLAST hits:"
head blast_results.txt

echo "Job completed successfully."