Merge pull request #8 from Australian-Structural-Biology-Computing/ad…

…d-dbs-variables-to-msa-pipeline Make pipeline work on UNSW Katana
Australian-Structural-Biology-Computing · Aug 8, 2024 · 632610b · 632610b
2 parents e676c33 + 03f2575
commit 632610b
Show file tree

Hide file tree

Showing 7 changed files with 34 additions and 21 deletions.
diff --git a/conf/base.config b/conf/base.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

diff --git a/conf/dbs.config b/conf/dbs.config
@@ -25,11 +25,13 @@ params {
     // Alphafold paths
     bfd_path               = "${params.alphafold2_db}/bfd/*"
     small_bfd_path         = "${params.alphafold2_db}/small_bfd/*"
-    alphafold2_params_path = "${params.alphafold2_db}/alphafold_params_*/*"
+    //alphafold2_params_path = "${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/**"
-    uniclust30_path        = "${params.alphafold2_db}/uniclust30/**"
+    //uniclust30_path        = "${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           = "${params.alphafold2_db}/uniprot/*"
@@ -40,11 +42,17 @@ params {
     mgnify_variable            = "${params.alphafold2_db}/mgnify/"
     pdb70_variable             = "${params.alphafold2_db}/pdb70/"
     pdb_mmcif_variable         = "${params.alphafold2_db}/pdb_mmcif/"
-    uniclust30_variable        = "${params.alphafold2_db}/uniclust30/"
+    //uniclust30_variable        = "${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/"
 
+    // 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"
+
     // Colabfold links
     colabfold_db_link = 'http://wwwuser.gwdg.de/~compbiol/colabfold/colabfold_envdb_202108.tar.gz'
     uniref30          = 'https://wwwuser.gwdg.de/~compbiol/colabfold/uniref30_2202.tar.gz'

diff --git a/modules/local/run_alphafold2_msa.nf b/modules/local/run_alphafold2_msa.nf
@@ -4,6 +4,7 @@
 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' :
@@ -34,28 +35,30 @@ process RUN_ALPHAFOLD2_MSA {
 
     script:
     def args = task.ext.args ?: ''
-    def db_preset = db_preset ? "full_dbs --bfd_database_path=./bfd/bfd_metaclust_clu_complete_id30_c90_final_seq.sorted_opt --uniclust30_database_path=./uniclust30/uniclust30_2018_08/uniclust30_2018_08" :
-        "reduced_dbs --small_bfd_database_path=./small_bfd/bfd-first_non_consensus_sequences.fasta"
+    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=${params.small_bfd_path}bfd-first_non_consensus_sequences.fasta"
     if (alphafold2_model_preset == 'multimer') {
-        alphafold2_model_preset += " --pdb_seqres_database_path=./pdb_seqres/pdb_seqres.txt --uniprot_database_path=./uniprot/uniprot.fasta "
+        alphafold2_model_preset += " --pdb_seqres_database_path=${params.pdb_seqres_variable}pdb_seqres.txt --uniprot_database_path=${params.uniprot_variable}/uniprot.fasta "
     }
     else {
-        alphafold2_model_preset += " --pdb70_database_path=./pdb70/pdb70_from_mmcif_200916/pdb70 "
+        alphafold2_model_preset += " --pdb70_database_path=${params.pdb70_variable}pdb70 "
     }
     """
-    if [ -f pdb_seqres/pdb_seqres.txt ]
-        then sed -i "/^\\w*0/d" pdb_seqres/pdb_seqres.txt
-    fi
+    #if [ -f pdb_seqres/pdb_seqres.txt ]
+    #    \$PDB_SEQRES_TEMP=\$(mktemp --directory)
+    #    cp ${params.pdb_seqres_variable}pdb_seqres.txt \${PDB_SEQRES_TEMP}/
+    #    then sed -i "/^\\w*0/d" \$PDB_SEQERS_TEMP/pdb_seqres.txt
+    #fi
     python3 /app/alphafold/run_msa.py \
         --fasta_paths=${fasta} \
         --model_preset=${alphafold2_model_preset} \
         --db_preset=${db_preset} \
         --output_dir=\$PWD \
         --data_dir=\$PWD \
-        --uniref90_database_path=./uniref90/uniref90.fasta \
-        --mgnify_database_path=./mgnify/mgy_clusters_2018_12.fa \
-        --template_mmcif_dir=./pdb_mmcif/mmcif_files \
-        --obsolete_pdbs_path=./pdb_mmcif/obsolete.dat  \
+        --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}  \
         $args
 
     cp "${fasta.baseName}"/features.pkl ./"${fasta.baseName}".features.pkl

diff --git a/modules/local/run_alphafold2_pred.nf b/modules/local/run_alphafold2_pred.nf
@@ -8,6 +8,7 @@ process RUN_ALPHAFOLD2_PRED {
     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'
 
     input:
     tuple val(meta), path(fasta)
@@ -36,7 +37,8 @@ process RUN_ALPHAFOLD2_PRED {
     script:
     def args = task.ext.args ?: ''
     """
-    if [ -d params/alphafold_params_* ]; then ln -r -s params/alphafold_params_*/* params/; fi
+    echo \$PWD
+    #if [ -d params/alphafold_params_* ]; then ln -r -s params/alphafold_params_*/* params/; fi
     python3 /app/alphafold/run_predict.py \
         --fasta_paths=${fasta} \
         --model_preset=${alphafold2_model_preset} \

diff --git a/pf_files/proteinfold_test.sh b/pf_files/proteinfold_test.sh
@@ -3,12 +3,13 @@ module load nextflow/23.04.4 java/11
 export SINGULARITY_CACHE_DIR=/srv/scratch/sbf/singularity_cache
 export NXF_SINGULARITY_CACHEDIR=/srv/scratch/sbf/singularity_cache
 
-nextflow run /srv/scratch/z5378336/proteinfold/main.nf \
+nextflow run ../main.nf \
     --input samplesheet.csv \
     --outdir test_out \
     --mode alphafold2 \
     --alphafold2_db /data/bio/alphafold \
     --full_dbs true \
     --alphafold2_model_preset monomer \
+    --alphafold2_mode 'split_msa_prediction' \
     --use_gpu true \
-    -profile singularity
+    -profile singularity \ 
diff --git a/pf_files/samplesheet.csv b/pf_files/samplesheet.csv
@@ -1,3 +1,2 @@
 sequence,fasta
-T1024,https://raw.githubusercontent.com/nf-core/test-datasets/proteinfold/testdata/sequences/T1024.fasta
-T1026,https://raw.githubusercontent.com/nf-core/test-datasets/proteinfold/testdata/sequences/T1026.fasta
+1L2Y,/srv/scratch/sbf/nextflow_pipelines-dev/proteinfold/pf_files/1L2Y.fasta
diff --git a/pf_files/test_out/pipeline_info/samplesheet.valid.csv b/pf_files/test_out/pipeline_info/samplesheet.valid.csv
@@ -1,2 +1,2 @@
 sequence,fasta
-1L2Y_T1,./1L2Y.fasta
+1L2Y_T1,/srv/scratch/sbf/nextflow_pipelines-dev/proteinfold/pf_files/1L2Y.fasta