#!/bin/bash #SBATCH --job-name=Beauveria_complete_pipeline #SBATCH --nodes=1 #SBATCH --ntasks-per-node=1 #SBATCH --cpus-per-task=128 #SBATCH --output=logs/pipeline_%j.out #SBATCH --error=logs/pipeline_%j.err #SBATCH --time=7-00:00:00 # ------------------------------------------------------------ # Beauveria bassiana multi‑genome analysis pipeline # Combines: download, preprocessing, BUSCO, BRAKER3, # funannotate, antiSMASH, BiG‑SCAPE, phylogenomics # ------------------------------------------------------------ set -euo pipefail trap 'echo "Error at line $LINENO" >&2' ERR # ---------------------------------------------------------------------- # 0. Configuration & directories # ---------------------------------------------------------------------- WORKDIR=$(pwd) mkdir -p logs raw_genome reference protein_dbs antismash_input antismash_output bigscape_input bigscape_results orthofinder_input orthofinder_output THREADS=$SLURM_CPUS_PER_TASK # Conda initialisation source $HOME/miniconda3/etc/profile.d/conda.sh # ---------------------------------------------------------------------- # 1. Download genomes and reference proteins (if not already present) # ---------------------------------------------------------------------- echo "[1] Downloading Beauveria genomes and reference proteins" conda activate ncbi_datasets if [ ! -d raw_genome ] || [ -z "$(ls -A raw_genome/*.fna 2>/dev/null)" ]; then datasets download genome taxon "Beauveria" --include genome --filename beauveria_genomic.zip unzip -q beauveria_genomic.zip -d temp_download find temp_download -name "*.fna" -exec mv {} raw_genome/ \; rm -rf temp_download beauveria_genomic.zip fi if [ ! -f reference/protein.faa ]; then mkdir -p reference datasets download genome accession GCF_000280675.1 --include protein --filename ref_prot.zip unzip -q ref_prot.zip -d ref_temp find ref_temp -name "*.faa" -exec cp {} reference/protein.faa \; rm -rf ref_temp ref_prot.zip fi conda deactivate # ---------------------------------------------------------------------- # 2. Preprocess genomes (clean, sort, mask) with funannotate # ---------------------------------------------------------------------- echo "[2] Preprocessing genomes with funannotate" conda activate funannotate_env cd raw_genome for f in *.fna; do base=${f%.fna} if [ ! -f "${base}.clean.sort.mask.fna" ]; then funannotate clean -i "$f" -o "${base}.clean.fna" funannotate sort -i "${base}.clean.fna" -o "${base}.clean.sort.fna" funannotate mask -i "${base}.clean.sort.fna" -o "${base}.clean.sort.mask.fna" fi done cd .. conda deactivate # ---------------------------------------------------------------------- # 3. BUSCO quality assessment (optional, but informative) # ---------------------------------------------------------------------- echo "[3] Running BUSCO quality assessment" conda activate busco_env cd raw_genome for genome in *.clean.sort.mask.fna; do base=${genome%.clean.sort.mask.fna} if [ ! -d "${base}_busco" ]; then busco -i "$genome" -o "${base}_busco" -l hypocreales_odb10 -m genome --cpu $THREADS --download_path ../busco_downloads fi done cd .. conda deactivate # ---------------------------------------------------------------------- # 4. Gene prediction with BRAKER3 using reference proteins # ---------------------------------------------------------------------- echo "[4] Running BRAKER3 gene prediction" conda activate braker3_env cd raw_genome REF_PROT="../reference/protein.faa" for genome in *.clean.sort.mask.fna; do base=${genome%.clean.sort.mask.fna} if [ ! -d "${base}_braker" ]; then braker.py \ --genome="$genome" \ --prot_seq="$REF_PROT" \ --workingdir="${base}_braker" \ --threads=$THREADS \ --gff3 \ --fungus \ --softmasking fi done cd .. conda deactivate # ---------------------------------------------------------------------- # 5. Functional annotation with funannotate (using BRAKER GFF3) # ---------------------------------------------------------------------- echo "[5] Running funannotate annotation" conda activate funannotate_env cd raw_genome for genome in *.clean.sort.mask.fna; do base=${genome%.clean.sort.mask.fna} gff="${base}_braker/braker.gff3" if [ -f "$gff" ] && [ ! -d "$base" ]; then funannotate annotate \ --fasta "$genome" \ --gff "$gff" \ -o "$base" \ --species "Beauveria bassiana" \ --cpus $THREADS fi done cd .. conda deactivate # ---------------------------------------------------------------------- # 6. antiSMASH detection of secondary metabolite clusters # ---------------------------------------------------------------------- echo "[6] Running antiSMASH" conda activate antismash_env # Collect all GenBank files from funannotate outputs find raw_genome -path "*/annotate_results/*.gbk" -exec cp {} antismash_input/ \; cd antismash_input for gbk in *.gbk; do base=${gbk%.gbk} if [ ! -d "../antismash_output/$base" ]; then antismash "$gbk" \ --taxon fungi \ --genefinding-tool none \ --output-dir ../antismash_output/$base \ --cpus $THREADS fi done cd .. conda deactivate # ---------------------------------------------------------------------- # 7. BiG‑SCAPE clustering of biosynthetic gene clusters # ---------------------------------------------------------------------- echo "[7] Running BiG‑SCAPE" conda activate bigscape_env # Copy all region GBK files to input directory find antismash_output -name "*region*.gbk" -exec cp {} bigscape_input/ \; if [ ! -d bigscape_results ] || [ -z "$(ls -A bigscape_results)" ]; then bigscape.py \ --inputdir bigscape_input \ --outputdir bigscape_results \ --threads $THREADS \ --mibig \ --mix \ --keep_singletons fi # Create final bigscape_output with cleaned results mkdir -p bigscape_output cp -r bigscape_results/networkfiles/*_hybrids bigscape_output/ find bigscape_output -type d -name "GCF_trees" -exec rm -rf {} + 2>/dev/null || true conda deactivate # ---------------------------------------------------------------------- # 8. BUSCO for phylogenomics (nucleotide mode, Metaeuk) # ---------------------------------------------------------------------- echo "[8] Running BUSCO for phylogenomics" conda activate busco_env mkdir -p busco_nt_output cd raw_genome for genome in *.fna; do base=${genome%.fna} if [ ! -d "../busco_nt_output/${base}_busco_nt" ]; then busco \ -i "$genome" \ -o "../busco_nt_output/${base}_busco_nt" \ -l hypocreales_odb10 \ -m genome \ --nt \ --metaeuk \ --cpu $THREADS fi done cd .. conda deactivate # ---------------------------------------------------------------------- # 9. Generate supermatrix from BUSCO single-copy orthologs # ---------------------------------------------------------------------- echo "[9] Generating supermatrix" conda activate busco_env # The custom script BUSCO_phylogenomics.py is assumed to be in PATH if command -v BUSCO_phylogenomics.py >/dev/null 2>&1; then BUSCO_phylogenomics.py \ -i raw_genome \ -o Bb_Phylogeny \ -t $THREADS \ -l hypocreales_odb10 \ -p 80 \ --supermatrix \ --nt else echo "ERROR: BUSCO_phylogenomics.py not found. Please ensure it is installed and in PATH." exit 1 fi conda deactivate # ---------------------------------------------------------------------- # 10. IQ‑TREE phylogenetic inference # ---------------------------------------------------------------------- echo "[10] Running IQ‑TREE" conda activate iqtree_env # Assumes iqtree environment exists cd Bb_Phylogeny if [ -f SUPERMATRIX.phylip ] && [ -f SUPERMATRIX.partitions.nex ]; then iqtree2 \ -s SUPERMATRIX.phylip \ -p SUPERMATRIX.partitions.nex \ -m MFP+MERGE \ -B 1000 \ --alrt 1000 \ -T AUTO else echo "ERROR: Supermatrix files not found. Check BUSCO supermatrix step." exit 1 fi conda deactivate echo "Pipeline completed at $(date)"