Clean up code

conf/dbs.config

@@ -21,37 +21,72 @@ params {
     pdb_seqres        = 'ftp://ftp.wwpdb.org/pub/pdb/derived_data/pdb_seqres.txt'
     uniprot_sprot     = 'ftp://ftp.ebi.ac.uk/pub/databases/uniprot/current_release/knowledgebase/complete/uniprot_sprot.fasta.gz'
     uniprot_trembl    = 'ftp://ftp.ebi.ac.uk/pub/databases/uniprot/current_release/knowledgebase/complete/uniprot_trembl.fasta.gz'
+
+    bfd_name           = params.bfd_prefix ?: 'bfd'
+    smallbfd_name      = params.smallbfd_prefix ?: 'smallbfd'
+    mgnify_name        = params.mgnify_prefix ?: 'mgnify'
+    pdb70_name         = params.pdb70_prefix ?: 'pdb70'
+    pdb_mmcif_name     = params.pdb_mmcif_prefix ?: 'pdb_mmcif'
+    uniclust30_name    = params.uniclust30_prefix ?: 'uniclust30'
+    uniref90_name      = params.uniref90_prefix ?: 'uniref90'
+    pdb_seqres_name    = params.pdb_seq_prefix ?: 'pdb_seqres'
+    uniprot_name       = params.uniprot_prefix ?: 'uniprot'
+    alphafold_params_name = params.alphafold_params_prefix ?: 'params/alphafold_params_*'
+    mmcif_files_name   = params.mmcif_path ?: 'pdb_mmcif/mmcif_files/'
+    mmcif_obsolete_name = params.mmcif_obsolete ?: 'pdb_mmcif/obsolete.dat'
+
+    uniclust30_db_name = params.uniclust30_db ?: 'uniclust30_2018_08'
+    bfd_first_non_consensus_sequences_name = params.bfd_first_non_consensus_sequences ?: 'bfd-first_non_consensus_sequences.fasta' 
+    uniprot_fasta_name = params.uniprot_fasta ?: 'uniprot.fasta'
+    pdb_seqres_txt_name = params.pdb_seqres_txt ?: 'pdb_seqres.txt'
+    bfd_metaclust_clu_complete_id30_c90_final_seq_sorted_opt_name = params.bfd_metaclust_clu_complete_id30_c90_final_seq_sorted_opt ?: 'bfd_metaclust_clu_complete_id30_c90_final_seq.sorted_opt'
+    uniref90_fasta_name = params.uniref90_fasta ?: 'uniref90.fasta'
+    mgy_clusters_fasta_name = params.mgy_clusters_fasta ?: 'mgy_clusters_2022_05.fa'
+
 
     // Alphafold paths
-    bfd_path               = "${params.alphafold2_db}/bfd/*"
-    small_bfd_path         = "${params.alphafold2_db}/small_bfd/*"
+    bfd_path               = "${params.alphafold2_db}/${bfd_name}/*"
+    small_bfd_path         = "${params.alphafold2_db}/${smallbfd_name}/*"
     //alphafold2_params_path = "${params.alphafold2_db}/alphafold_params_*/*"
-    alphafold2_params_path = "${params.alphafold2_db}/params/*"
-    mgnify_path            = "${params.alphafold2_db}/mgnify/*"
-    pdb70_path             = "${params.alphafold2_db}/pdb70/**"
-    pdb_mmcif_path         = "${params.alphafold2_db}/pdb_mmcif/**"
+    //alphafold2_params_path = "${params.alphafold2_db}/params/*"
+    alphafold2_params_path = "${params.alphafold2_db}/${alphafold_params_name}/*"
+    mgnify_path            = "${params.alphafold2_db}/${mgnify_name}/*"
+    pdb70_path             = "${params.alphafold2_db}/${pdb70_name}/**"
+    pdb_mmcif_path         = "${params.alphafold2_db}/${pdb_mmcif_name}/**"
     //uniclust30_path        = "${params.alphafold2_db}/uniclust30/**"
-    uniclust30_path        = "/srv/scratch/sbf/uniclust30/**"
-    uniref90_path          = "${params.alphafold2_db}/uniref90/*"
-    pdb_seqres_path        = "${params.alphafold2_db}/pdb_seqres/*"
-    uniprot_path           = "${params.alphafold2_db}/uniprot/*"
+    //uniclust30_path        = "/srv/scratch/sbf/uniclust30/**"
+    uniclust30_path        = "${params.alphafold2_db}/${uniclust30_name}/**"
+    uniref90_path          = "${params.alphafold2_db}/${uniref90_name}/*"
+    pdb_seqres_path        = "${params.alphafold2_db}/${pdb_seqres_name}/*"
+    uniprot_path           = "${params.alphafold2_db}/${uniprot_name}/*"
 
     // Alphafold variables
-    bfd_variable               = "${params.alphafold2_db}/bfd/"
-    small_bfd_variable         = "${params.alphafold2_db}/smallbfd/"
-    mgnify_variable            = "${params.alphafold2_db}/mgnify/"
-    pdb70_variable             = "${params.alphafold2_db}/pdb70/"
-    pdb_mmcif_variable         = "${params.alphafold2_db}/pdb_mmcif/"
+    //bfd_variable               = "${params.alphafold2_db}/bfd/"
+    bfd_dir_path               = "${params.alphafold2_db}/${bfd_name}/"
+    //small_bfd_variable         = "${params.alphafold2_db}/smallbfd/"
+    small_bfd_dir_path         = "${params.alphafold2_db}/${smallbfd_name}/"
+    //mgnify_variable            = "${params.alphafold2_db}/mgnify/"
+    mgnify_dir_path            = "${params.alphafold2_db}/${mgnify_name}/"
+    //pdb70_variable             = "${params.alphafold2_db}/pdb70/"
+    pdb70_dir_path             = "${params.alphafold2_db}/${pdb70_name}/"
+    //pdb_mmcif_variable         = "${params.alphafold2_db}/pdb_mmcif/"
+    pdb_mmcif_dir_path         = "${params.alphafold2_db}/${pdb_mmcif_name}/"
     //uniclust30_variable        = "${params.alphafold2_db}/uniclust30/"
-    uniclust30_variable        = "/srv/scratch/sbf/uniclust30/"
-    uniref90_variable          = "${params.alphafold2_db}/uniref90/"
-    pdb_seqres_variable        = "${params.alphafold2_db}/pdb_seqres/"
-    uniprot_variable           = "${params.alphafold2_db}/uniprot/"
+    //uniclust30_variable        = "/srv/scratch/sbf/uniclust30/"
+    uniclust30_dir_path        = "${params.alphafold2_db}/${uniclust30_name}/"
+    //uniref90_variable          = "${params.alphafold2_db}/uniref90/"
+    uniref90_dir_path          = "${params.alphafold2_db}/${uniref90_name}/"
+    //pdb_seqres_variable        = "${params.alphafold2_db}/pdb_seqres/"
+    pdb_seqres_dir_path        = "${params.alphafold2_db}/${pdb_seqres_name}/"
+    //uniprot_variable           = "${params.alphafold2_db}/uniprot/"
+    uniprot_dir_path           = "${params.alphafold2_db}/${uniprot_name}/"
 
     // Alphafold MSA Variables
-    mgnify_database_path       = "${params.alphafold2_db}/mgnify/"
-    template_mmcif_dir         = "${params.alphafold2_db}/pdb_mmcif/mmcif_files/"
-    obsolete_pdbs_path         = "${params.alphafold2_db}/pdb_mmcif/obsolete.dat"
+    mgnify_database_path       = "${params.alphafold2_db}/${mgnify_name}/"
+    //template_mmcif_dir         = "${params.alphafold2_db}/pdb_mmcif/mmcif_files/"
+    template_mmcif_dir         = "${params.alphafold2_db}/${mmcif_files_name}/"
+    //obsolete_pdbs_path         = "${params.alphafold2_db}/pdb_mmcif/obsolete.dat"
+    obsolete_pdbs_path         = "${params.alphafold2_db}/${mmcif_obsolete_name}"
 
     // Colabfold links
     colabfold_db_link = 'http://wwwuser.gwdg.de/~compbiol/colabfold/colabfold_envdb_202108.tar.gz'

conf/katana.config

@@ -19,7 +19,7 @@ process {
     memory = { check_max( 6.GB * task.attempt, 'memory' ) }
     time   = { check_max( 4.h  * task.attempt, 'time'   ) }
 
-    executor = 'pbspro'
+    //executor = 'pbspro'
 
     errorStrategy = { task.exitStatus in [143,137,104,134,139] ? 'retry' : 'finish' }
     maxRetries    = 1

modules/local/run_alphafold2.nf

@@ -5,9 +5,12 @@ process RUN_ALPHAFOLD2 {
     tag "$meta.id"
     label 'process_medium'
 
-    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
-        'docker://nfcore/proteinfold_alphafold2_standard:1.0.0' :
-        'nfcore/proteinfold_alphafold2_standard:1.0.0' }"
+    // Exit if running this module with -profile conda / -profile mamba
+    if (workflow.profile.tokenize(',').intersect(['conda', 'mamba']).size() >= 1) {
+        error("Local RUN_ALPHAFOLD2 module does not support Conda. Please use Docker / Singularity / Podman instead.")
+    }
+
+    container "nf-core/proteinfold_alphafold2_standard:1.1.1"
 
     input:
     tuple val(meta), path(fasta)
@@ -19,7 +22,7 @@ process RUN_ALPHAFOLD2 {
     path ('mgnify/*')
     path ('pdb70/*')
     path ('pdb_mmcif/*')
-    path ('uniclust30/*')
+    path ('uniref30/*')
     path ('uniref90/*')
     path ('pdb_seqres/*')
     path ('uniprot/*')
@@ -34,29 +37,35 @@ process RUN_ALPHAFOLD2 {
 
     script:
     def args = task.ext.args ?: ''
-    def db_preset = db_preset ? "full_dbs --bfd_database_path=${params.bfd_variable}bfd_metaclust_clu_complete_id30_c90_final_seq.sorted_opt --uniclust30_database_path=${params.uniclust30_variable}uniclust30_2018_08/uniclust30_2018_08" :
-        "reduced_dbs --small_bfd_database_path=${params.small_bfd_path}bfd-first_non_consensus_sequences.fasta"
+    def db_preset = db_preset ? "full_dbs --bfd_database_path=${params.alphafold2_db}/bfd/bfd_metaclust_clu_complete_id30_c90_final_seq.sorted_opt --uniref30_database_path=${params.alphafold2_db}/uniref30/UniRef30_2021_03" :
+        "reduced_dbs --small_bfd_database_path=${params.alphafold2_db}/small_bfd/bfd-first_non_consensus_sequences.fasta"
     if (alphafold2_model_preset == 'multimer') {
-        alphafold2_model_preset += " --pdb_seqres_database_path=${params.pdb_seqres_variable}pdb_seqres.txt --uniprot_database_path=${params.uniprot_variable}uniprot.fasta "
+        alphafold2_model_preset += " --pdb_seqres_database_path=${params.alphafold2_db}/pdb_seqres/pdb_seqres.txt --uniprot_database_path=${params.alphafold2_db}/uniprot/uniprot.fasta "
     }
     else {
-        alphafold2_model_preset += " --pdb70_database_path=${params.pdb70_variable}pdb70 "
+        alphafold2_model_preset += " --pdb70_database_path=${params.alphafold2_db}/pdb70/pdb70_from_mmcif_200916/pdb70 "
     }
     """
-    if [ -f ${params.pdb_seqres_variable}pdb_seqres.txt ]
-        then sed -i "/^\\w*0/d" ${params.pdb_seqres_variable}pdb_seqres.txt
+    RUNTIME_TMP=\$(mktemp -d)
+    nvcc --version 2>&1 | tee /home/z3545907/nvcc.txt
+    nvidia-smi 2>&1 | tee /home/z3545907/nvidia-smi.txt
+    if [ -f ${params.alphafold2_db}/pdb_seqres/pdb_seqres.txt ]
+        cp ${params.alphafold2_db}/pdb_seqres/pdb_seqres.txt \${RUNTIME_TMP}
+        then sed -i "/^\\w*0/d" \${RUNTIME_TMP}/pdb_seqres.txt
     fi
-    if [ -d params/alphafold_params_* ]; then ln -r -s params/alphafold_params_*/* params/; fi
+    if [ -d ${params.alphafold2_db}/params/ ]; then ln -r -s ${params.alphafold2_db}/params params; fi
     python3 /app/alphafold/run_alphafold.py \
         --fasta_paths=${fasta} \
         --model_preset=${alphafold2_model_preset} \
         --db_preset=${db_preset} \
         --output_dir=\$PWD \
         --data_dir=\$PWD \
-        --uniref90_database_path=${params.uniref90_variable}uniref90.fasta \
-        --template_mmcif_dir=${params.pdb_mmcif_variable}mmcif_files \
-        --obsolete_pdbs_path=${params.pdb_mmcif_variable}obsolete.dat \
+        --uniref90_database_path=${params.alphafold2_db}/uniref90/uniref90.fasta \
+        --mgnify_database_path=${params.alphafold2_db}/mgnify/mgy_clusters_2022_05.fa \
+        --template_mmcif_dir=${params.alphafold2_db}/pdb_mmcif/mmcif_files \
+        --obsolete_pdbs_path=${params.alphafold2_db}/pdb_mmcif/obsolete.dat \
         --random_seed=53343 \
+        --use_gpu_relax \
         $args
 
     cp "${fasta.baseName}"/ranked_0.pdb ./"${fasta.baseName}".alphafold.pdb
@@ -69,6 +78,7 @@ process RUN_ALPHAFOLD2 {
     echo -e Positions"\\t"rank_0"\\t"rank_1"\\t"rank_2"\\t"rank_3"\\t"rank_4 > header.tsv
     cat header.tsv plddt.tsv > ../"${fasta.baseName}"_plddt_mqc.tsv
     cd ..
+    rm -rf "\${RUNTIME_TMP}"
 
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":

modules/local/run_alphafold2_msa.nf

@@ -4,11 +4,13 @@
 process RUN_ALPHAFOLD2_MSA {
     tag   "$meta.id"
     label 'process_medium'
-    debug true
 
-    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
-        'docker://nfcore/proteinfold_alphafold2_msa:1.0.0' :
-        'nfcore/proteinfold_alphafold2_msa:1.0.0' }"
+    // Exit if running this module with -profile conda / -profile mamba
+    if (workflow.profile.tokenize(',').intersect(['conda', 'mamba']).size() >= 1) {
+        error("Local RUN_ALPHAFOLD2_MSA module does not support Conda. Please use Docker / Singularity / Podman instead.")
+    }
+
+    container "nf-core/proteinfold_alphafold2_msa:1.1.1"
 
     input:
     tuple val(meta), path(fasta)
@@ -20,7 +22,7 @@ process RUN_ALPHAFOLD2_MSA {
     path ('mgnify/*')
     path ('pdb70/*')
     path ('pdb_mmcif/*')
-    path ('uniclust30/*')
+    path ('uniref30/*')
     path ('uniref90/*')
     path ('pdb_seqres/*')
     path ('uniprot/*')
@@ -35,33 +37,34 @@ process RUN_ALPHAFOLD2_MSA {
 
     script:
     def args = task.ext.args ?: ''
-    def db_preset = db_preset ? "full_dbs --bfd_database_path=${params.bfd_variable}bfd_metaclust_clu_complete_id30_c90_final_seq.sorted_opt --uniclust30_database_path=${params.uniclust30_variable}uniclust30_2018_08" :
-        "reduced_dbs --small_bfd_database_path=${params.small_bfd_path}bfd-first_non_consensus_sequences.fasta"
+    def db_preset = db_preset ? "full_dbs --bfd_database_path=${params.alphafold2_db}/bfd/bfd_metaclust_clu_complete_id30_c90_final_seq.sorted_opt --uniref30_database_path=${params.alphafold2_db}/uniref30/UniRef30_2021_03" :
+        "reduced_dbs --small_bfd_database_path=${params.alphafold2_db}/small_bfd/bfd-first_non_consensus_sequences.fasta"
     if (alphafold2_model_preset == 'multimer') {
-        alphafold2_model_preset += " --pdb_seqres_database_path=${params.pdb_seqres_variable}pdb_seqres.txt --uniprot_database_path=${params.uniprot_variable}/uniprot.fasta "
+        alphafold2_model_preset += " --pdb_seqres_database_path=${params.alphafold2_db}/pdb_seqres/pdb_seqres.txt --uniprot_database_path=${params.alphafold2_db}/uniprot/uniprot.fasta "
     }
     else {
-        alphafold2_model_preset += " --pdb70_database_path=${params.pdb70_variable}pdb70 "
+        alphafold2_model_preset += " --pdb70_database_path=${params.alphafold2_db}/pdb70/pdb70_from_mmcif_200916/pdb70 "
     }
     """
-    #if [ -f pdb_seqres/pdb_seqres.txt ]
-    #    \$PDB_SEQRES_TEMP=\$(mktemp --directory)
-    #    cp ${params.pdb_seqres_variable}pdb_seqres.txt \${PDB_SEQRES_TEMP}/
-    #    then sed -i "/^\\w*0/d" \$PDB_SEQERS_TEMP/pdb_seqres.txt
-    #fi
+    RUNTIME_TMP=\$(mktemp -d)
+    if [ -f ${params.alphafold2_db}/pdb_seqres/pdb_seqres.txt ]
+        cp ${params.alphafold2_db}/pdb_seqres/pdb_seqres.txt \${RUNTIME_TMP}
+        then sed -i "/^\\w*0/d" \${RUNTIME_TMP}/pdb_seqres.txt
+    fi
     python3 /app/alphafold/run_msa.py \
         --fasta_paths=${fasta} \
         --model_preset=${alphafold2_model_preset} \
         --db_preset=${db_preset} \
         --output_dir=\$PWD \
         --data_dir=\$PWD \
-        --uniref90_database_path=${params.uniref90_variable}uniref90.fasta \
-        --mgnify_database_path=${params.mgnify_database_path}/mgy_clusters_2022_05.fa \
-        --template_mmcif_dir=${params.template_mmcif_dir} \
-        --obsolete_pdbs_path=${params.obsolete_pdbs_path}  \
+        --uniref90_database_path=${params.alphafold2_db}/uniref90/uniref90.fasta \
+        --mgnify_database_path=${params.alphafold2_db}/mgnify/mgy_clusters_2022_05.fa \
+        --template_mmcif_dir=${params.alphafold2_db}/pdb_mmcif/mmcif_files \
+        --obsolete_pdbs_path=${params.alphafold2_db}/pdb_mmcif/obsolete.dat  \
         $args
 
     cp "${fasta.baseName}"/features.pkl ./"${fasta.baseName}".features.pkl
+    rm -rf "\${RUNTIME_TMP}"
 
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":

modules/local/run_alphafold2_pred.nf

@@ -3,12 +3,15 @@
  */
 process RUN_ALPHAFOLD2_PRED {
     tag   "$meta.id"
-    label 'process_medium', 'gpu_compute'
+    label 'process_medium'
+    label 'gpu_compute'
 
-    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
-        'docker://nfcore/proteinfold_alphafold2_split:1.0.0' :
-        'nfcore/proteinfold_alphafold2_split:1.0.0' }"
-    echo 'true'
+    // Exit if running this module with -profile conda / -profile mamba
+    if (workflow.profile.tokenize(',').intersect(['conda', 'mamba']).size() >= 1) {
+        error("Local RUN_ALPHAFOLD2_PRED module does not support Conda. Please use Docker / Singularity / Podman instead.")
+    }
+
+    container "nf-core/proteinfold_alphafold2_split:1.1.1"
 
     input:
     tuple val(meta), path(fasta)
@@ -20,7 +23,7 @@ process RUN_ALPHAFOLD2_PRED {
     path ('mgnify/*')
     path ('pdb70/*')
     path ('pdb_mmcif/*')
-    path ('uniclust30/*')
+    path ('uniref30/*')
     path ('uniref90/*')
     path ('pdb_seqres/*')
     path ('uniprot/*')
@@ -37,15 +40,15 @@ process RUN_ALPHAFOLD2_PRED {
     script:
     def args = task.ext.args ?: ''
     """
-    echo \$PWD
-    #if [ -d params/alphafold_params_* ]; then ln -r -s params/alphafold_params_*/* params/; fi
+    if [ -d ${params.alphafold2_db}/params/ ]; then ln -r -s ${params.alphafold2_db}/params params; fi
     python3 /app/alphafold/run_predict.py \
         --fasta_paths=${fasta} \
         --model_preset=${alphafold2_model_preset} \
         --output_dir=\$PWD \
         --data_dir=\$PWD \
         --random_seed=53343 \
         --msa_path=${msa} \
+        --use_gpu_relax \
         $args
 
     cp "${fasta.baseName}"/ranked_0.pdb ./"${fasta.baseName}".alphafold.pdb

nextflow.config

@@ -20,6 +20,48 @@ params {
     full_dbs                    =  false // true full_dbs, false reduced_dbs
     alphafold2_model_preset     = "monomer" // for AF2 {monomer (default), monomer_casp14, monomer_ptm, multimer}
     alphafold2_db               = null
+
+    // Database prefixes
+    bfd_prefix			= null
+    smallbfd_prefix		= null
+    mgnify_prefix		= null
+    pdb70_prefix		= null
+    pdb_mmcif_prefix		= null
+    uniclust30_prefix		= null
+    uniref90_prefix		= null
+    pdb_seq_prefix		= null
+    uniprot_prefix		= null
+    alphafold_params_prefix	= null
+    mmcif_path			= null
+    mmcif_obsolete		= null
+    uniclust30_db		= null
+    bfd_first_non_consensus_sequences = null
+    uniprot_fasta		= null
+    pdb_seqres_txt		= null
+    bfd_metaclust_clu_complete_id30_c90_final_seq_sorted_opt = null
+    uniref90_fasta		= null
+    mgy_clusters_fasta		= null
+    uniclust30_prefix		= null
+
+    bfd_name			= null
+    smallbfd_name		= null
+    mgnify_name			= null
+    pdb70_name			= null
+    pdb_mmcif_name		= null
+    uniclust30_name		= null
+    uniref90_name		= null
+    pdb_seqres_name		= null
+    uniprot_name		= null
+    alphafold_params_name	= null
+    mmcif_files_name		= null
+    mmcif_obsolete_name		= null
+    uniclust30_db_name		= null
+    bfd_first_non_consensus_sequences_name = null
+    uniprot_fasta_name		= null
+    pdb_seqres_txt_name		= null
+    bfd_metaclust_clu_complete_id30_c90_final_seq_sorted_opt_name = null
+    uniref90_fasta_name		= null
+    mgy_clusters_fasta_name	= null
 
     // Alphafold2 links
     bfd                         = null
@@ -104,8 +146,10 @@ params {
 
 }
 
+spack.enabled = true
+
 // Load base.config by default for all pipelines
-includeConfig 'conf/base.config'
+includeConfig 'conf/katana.config'
 
 // Load nf-core custom profiles from different Institutions
 try {

pf_files/proteinfold_test.sh

@@ -1,15 +1,16 @@
-module load nextflow/23.04.4 java/11
+module load nextflow/23.04.4 java/11 cuda/11.8.0
 
-export SINGULARITY_CACHE_DIR=/srv/scratch/sbf/singularity_cache
-export NXF_SINGULARITY_CACHEDIR=/srv/scratch/sbf/singularity_cache
+export SINGULARITY_CACHE_DIR=/srv/scratch/$USER/Singularity/cache
+export NXF_SINGULARITY_CACHEDIR=/srv/scratch/$USER/Singularity/cache
 
 nextflow run ../main.nf \
     --input samplesheet.csv \
     --outdir test_out \
     --mode alphafold2 \
-    --alphafold2_db /data/bio/alphafold \
+    --alphafold2_db /mnt/af2/ \
     --full_dbs true \
-    --alphafold2_model_preset monomer \
+    --alphafold2_model_preset multimer \
+    --alphafold_params_name 'params' \
     --alphafold2_mode 'split_msa_prediction' \
     --use_gpu true \
     -profile singularity \