update nf pipeline. should output rhogs/pickle folder in nf, otherwis…

…e sbatch error .command.run too long
DessimozLab · Sep 3, 2023 · 64fc2a7 · 64fc2a7
1 parent ebce1c7
commit 64fc2a7
Show file tree

Hide file tree

Showing 7 changed files with 49 additions and 80 deletions.
diff --git a/FastOMA.nf b/FastOMA.nf
@@ -8,9 +8,8 @@ params.proteomes = params.proteome_folder + "/*"
 params.hogmap_in = params.input_folder + "/hogmap_in"
 
 params.hogmap_folder = params.output_folder + "/hogmap"
-//params.rhogs_folder = params.output_folder + "/rhogs_all"
 params.species_tree = params.input_folder + "/species_tree.nwk"
-//params.pickles_rhogs_folder = params.output_folder + "/pickle_rhogs"
+params.pickles_temp = params.output_folder + "/pickles_temp"
 params.genetrees_folder = params.output_folder + "/genetrees"
 
 
@@ -35,13 +34,13 @@ process omamer_run{
 }
 
 
-process infer_roothogs{   // publishDir  params.rhogs_folder
+process infer_roothogs{
   input:
     val ready_omamer_run
     path hogmap_folder
     path proteome_folder
   output:
-    path "rhogs_all" // path "rhogs_all/*"
+    path "omamer_rhogs"
     path "gene_id_dic_xml.pickle"
     val true         // nextflow-io.github.io/patterns/state-dependency/
   script:
@@ -51,11 +50,10 @@ process infer_roothogs{   // publishDir  params.rhogs_folder
 }
 
 
-process batch_roothogs{ // publishDir params.output_folder
+process batch_roothogs{
   input:
     val ready_infer_roothogs
-    //path rhogs_folder
-    path "rhogs_all"
+    path "omamer_rhogs"
   output:
     path "rhogs_rest/*", optional: true
     path "rhogs_big/*" , optional: true
@@ -66,15 +64,15 @@ process batch_roothogs{ // publishDir params.output_folder
     """
 }
 
-process hog_big{ //publishDir params.pickles_rhogs_folder
+process hog_big{
+  publishDir params.pickles_temp
   cpus  6
   time {20.h}     // for very big rhog it might need more, or you could re-run and add `-resume`
   memory {20.GB}
   input:
     val rhogsbig_tree_ready
   output:
-    path "pickle_rhogs/*.pickle"
-    // path "*.pickle"
+    path "*.pickle"
     path "*.fa", optional: true   // msa         if write True
     path "*.nwk", optional: true  // gene trees  if write True
     val true
@@ -84,13 +82,12 @@ process hog_big{ //publishDir params.pickles_rhogs_folder
     """
 }
 
-process hog_rest{ //publishDir params.pickles_rhogs_folder
+process hog_rest{
+  publishDir params.pickles_temp
   input:
     val rhogsrest_tree_ready
   output:
-    path "pickle_rhogs/*.pickle"
-    // path "*.pickle"
-    //path "pickle_rhogs/*.pickle"
+    path "*.pickle"
     path "*.fa" , optional: true   // msa         if write True
     path "*.nwk" , optional: true  // gene trees  if write True
     val true
@@ -106,10 +103,10 @@ process collect_subhogs{
   publishDir params.output_folder, mode: 'copy'
   input:
     val ready_hog_rest
-    val ready_hog_big     // path pickle_rhogs   // this is for depenedcy
-    path "pickle_rhogs/"  // "*.pickle"  // path "pickle_rhogs"   // this is the folder includes pickles_rhogs
+    val ready_hog_big
+    path "pickles_temp"   // this is the folder includes pickles_rhogs
     path "gene_id_dic_xml.pickle"
-    path "rhogs_all"
+    path "omamer_rhogs"
   output:
     path "output_hog.orthoxml"
     path "OrthologousGroupsFasta"
@@ -121,28 +118,25 @@ process collect_subhogs{
     """
 }
 
-
 workflow {
     proteomes = Channel.fromPath(params.proteomes,  type:'any' ,checkIfExists:true)
     proteome_folder = Channel.fromPath(params.proteome_folder)
     hogmap_folder = Channel.fromPath(params.hogmap_folder)
-//    rhogs_folder = Channel.fromPath(params.rhogs_folder)
 
     genetrees_folder = Channel.fromPath(params.genetrees_folder)
     hogmap_in = Channel.fromPath(params.hogmap_in)
 
-//    pickles_rhogs_folder =  Channel.fromPath(params.pickles_rhogs_folder)
-    omamerdb = Channel.fromPath(params.input_folder+"/omamerdb.h5")     // proteomes.view{"prot ${it}"}
+    pickles_temp =  Channel.fromPath(params.pickles_temp)
+    omamerdb = Channel.fromPath(params.input_folder+"/omamerdb.h5")
     proteomes_omamerdb = proteomes.combine(omamerdb)
     proteomes_omamerdb_inputhog = proteomes_omamerdb.combine(hogmap_in) // proteomes_omamerdb_inputhog.view{" rhogsbig ${it}"}
     (hogmap, ready_omamer_run)= omamer_run(proteomes_omamerdb_inputhog)
     ready_omamer_run_c = ready_omamer_run.collect()
 
-    (rhogs_folder, gene_id_dic_xml, ready_infer_roothogs) = infer_roothogs(ready_omamer_run_c, hogmap_folder, proteome_folder)
-    // rhogs_folder and "rhogs_all" are the same
+    (omamer_rhogs, gene_id_dic_xml, ready_infer_roothogs) = infer_roothogs(ready_omamer_run_c, hogmap_folder, proteome_folder)
     ready_infer_roothogs_c = ready_infer_roothogs.collect()
 
-    (rhogs_rest_list, rhogs_big_list, ready_batch_roothogs) = batch_roothogs(ready_infer_roothogs_c, rhogs_folder)
+    (rhogs_rest_list, rhogs_big_list, ready_batch_roothogs) = batch_roothogs(ready_infer_roothogs_c, omamer_rhogs)
     ready_batch_roothogs_c = ready_batch_roothogs.collect()
 
     species_tree = Channel.fromPath(params.species_tree)
@@ -155,16 +149,8 @@ workflow {
     rhogsrest_tree =  rhogsrest.combine(species_tree)
     rhogsrest_tree_ready = rhogsrest_tree.combine(ready_batch_roothogs_c)
     (pickle_rest_rhog,  msas_out_rest, genetrees_out_test, ready_hog_rest) = hog_rest(rhogsrest_tree_ready)
-    all_pickles = pickle_big_rhog.mix(pickle_rest_rhog).collect() // all_pickles.view() //     pickle_rhogs_folder = Channel.fromPath(params.output_folder+"/pickle_rhogs")
 
-    (orthoxml_file, OrthologousGroupsFasta, OrthologousGroups_tsv, rootHOGs_tsv)  = collect_subhogs(ready_hog_rest.collect(), ready_hog_big.collect(), all_pickles, gene_id_dic_xml, rhogs_folder)  // pickles_rhogs_folder
+    (orthoxml_file, OrthologousGroupsFasta, OrthologousGroups_tsv, rootHOGs_tsv)  = collect_subhogs(ready_hog_rest.collect(), ready_hog_big.collect(), pickles_temp, gene_id_dic_xml, omamer_rhogs)
     orthoxml_file.view{" output orthoxml file ${it}"}
 
 }
-
-// memory {12.GB * (2*task.attempt - 1)}
-//    time {24.hour}
-//    errorStrategy {
-//      task.exitStatus in [1,99,143,137,104,134,139,145,140] ? ‘retry’ : ‘terminate’
-//    }
-//    maxRetries 4
diff --git a/FastOMA/__init__.py b/FastOMA/__init__.py
@@ -1,3 +1,3 @@
 
 __packagename__ = "FastOMA"
-__version__ = "0.0.6"
+__version__ = "0.1.0"
diff --git a/FastOMA/batch_roothogs.py b/FastOMA/batch_roothogs.py
@@ -91,7 +91,7 @@ def folder_1h_rhog(address_rhogs_folder, output_folder_big, output_folder_rest):
 
 def batch_roothogs():
 
-    input_rhog = "./rhogs_all/" #
+    input_rhog = "./omamer_rhogs/" #
     output_folder_big = "./rhogs_big/"
     output_folder_rest = "./rhogs_rest/"
     folder_1h_rhog(input_rhog, output_folder_big, output_folder_rest)

diff --git a/FastOMA/collect_subhogs.py b/FastOMA/collect_subhogs.py
@@ -27,7 +27,7 @@ def collect_subhogs():
     # tr|A0A0N7KCI6|A0A0N7KCI6_ORYSJ
     # for qfo benchamrk, the middle should be wirtten in the file
 
-    pickle_folder = "./pickle_rhogs/" #pickle_rhogs
+    pickle_folder = "./pickles_temp/" #pickle_rhogs
     output_xml_name = "./output_hog.orthoxml"
     gene_id_pickle_file = "./gene_id_dic_xml.pickle"
 
@@ -136,8 +136,8 @@ def max_og_tree(tree):
         return og_prot_list
 
     input_orthoxml = output_xml_name # sys.argv[1]  # "out_folder/output_hog_.orthoxml"
-    rhog_all_folder = "./rhogs_all/" #sys.argv[2] + "/"  # "out_folder/rhogs_all/"
-    fasta_format = "fa"  # of the rhogs_all
+    rhog_all_folder = "./omamer_rhogs/" #sys.argv[2] + "/"  # "out_folder/rhogs_all/"
+    fasta_format = "fa"  # of the rhogs
 
     output_file_og_tsv = "OrthologousGroups.tsv"
 
@@ -168,12 +168,12 @@ def max_og_tree(tree):
     for hog_id, og_prot_list in OGs.items():  # hog_id="HOG_0667494_sub10524"
         rhog_id = "_".join(hog_id.split("_")[:2])
 
-        rhogs_all_address = rhog_all_folder + rhog_id + "." + fasta_format
-        rhogs_all_prots = list(SeqIO.parse(rhogs_all_address, "fasta"))
+        omamer_rhogs_all_address = rhog_all_folder + rhog_id + "." + fasta_format
+        omamer_rhogs_all_prots = list(SeqIO.parse(omamer_rhogs_all_address, "fasta"))
 
         og_prots = []
         og_prot_list = OGs[hog_id]
-        for rhogs_prot in rhogs_all_prots:
+        for rhogs_prot in omamer_rhogs_all_prots:
             if rhogs_prot.id.split("||")[0] in og_prot_list:
                 sp = rhogs_prot.id.split("||")[1]
                 rhogs_prot.description += " [" + sp + "]"

diff --git a/FastOMA/infer_roothogs.py b/FastOMA/infer_roothogs.py
@@ -57,11 +57,11 @@ def infer_roothogs():
                                                                                  query_species_names,
                                                                                  query_prot_names_species_mapped)
     # for pure usage of this python file, you can set the output folder
-    # output_folder_rhog = _config.in_folder + "rhogs_all/"
+    # output_folder_rhog = _config.in_folder + "rhogs_all/" // omamer_rhogs
     # using nextflow
 
 
 
     # import sys
-    output_folder_rhog = "./rhogs_all/" #  sys.argv[1]  #
+    output_folder_rhog = "./omamer_rhogs/" #  sys.argv[1]  #
     rhogid_num_list_filt1 = _utils_roothog.write_rhog(rhogids_list_filt, rhogids_prot_records_query_filt, output_folder_rhog, 2)  # min_rhog_size, max_rhog_size
diff --git a/FastOMA/infer_subhogs.py b/FastOMA/infer_subhogs.py
@@ -53,7 +53,7 @@ def infer_subhogs():
     if inferhog_concurrent_on:
         print("parallelization for subhog inference is on.")
 
-    pickles_rhog_folder = "./pickle_rhogs/"
+    pickles_rhog_folder = "./" # pickles_temp/ pickle_rhogs
     if not os.path.exists(pickles_rhog_folder):
         os.makedirs(pickles_rhog_folder)
 

diff --git a/FastOMA_light.nf b/FastOMA_light.nf
@@ -8,15 +8,13 @@ params.proteomes = params.proteome_folder + "/*"
 params.hogmap_in = params.input_folder + "/hogmap_in"
 
 params.hogmap_folder = params.output_folder + "/hogmap"
-//params.rhogs_folder = params.output_folder + "/rhogs_all"
 params.species_tree = params.input_folder + "/species_tree.nwk"
-//params.pickles_rhogs_folder = params.output_folder + "/pickle_rhogs"
+params.pickles_temp = params.output_folder + "/pickles_temp"
 params.genetrees_folder = params.output_folder + "/genetrees"
 
 
 process omamer_run{
   time {4.h}
-  memory {4.GB}
   publishDir params.hogmap_folder
   input:
     path proteomes_omamerdb_inputhog
@@ -35,13 +33,13 @@ process omamer_run{
 }
 
 
-process infer_roothogs{   // publishDir  params.rhogs_folder
+process infer_roothogs{
   input:
     val ready_omamer_run
     path hogmap_folder
     path proteome_folder
   output:
-    path "rhogs_all" // path "rhogs_all/*"
+    path "omamer_rhogs"
     path "gene_id_dic_xml.pickle"
     val true         // nextflow-io.github.io/patterns/state-dependency/
   script:
@@ -51,11 +49,10 @@ process infer_roothogs{   // publishDir  params.rhogs_folder
 }
 
 
-process batch_roothogs{ // publishDir params.output_folder
+process batch_roothogs{
   input:
     val ready_infer_roothogs
-    //path rhogs_folder
-    path "rhogs_all"
+    path "omamer_rhogs"
   output:
     path "rhogs_rest/*", optional: true
     path "rhogs_big/*" , optional: true
@@ -66,15 +63,14 @@ process batch_roothogs{ // publishDir params.output_folder
     """
 }
 
-process hog_big{ //publishDir params.pickles_rhogs_folder
+process hog_big{
+  publishDir params.pickles_temp
   cpus  2
   time {20.h}     // for very big rhog it might need more, or you could re-run and add `-resume`
-  memory {4.GB}
   input:
     val rhogsbig_tree_ready
   output:
-    path "pickle_rhogs/*.pickle"
-    // path "*.pickle"
+    path "*.pickle"
     path "*.fa", optional: true   // msa         if write True
     path "*.nwk", optional: true  // gene trees  if write True
     val true
@@ -84,13 +80,12 @@ process hog_big{ //publishDir params.pickles_rhogs_folder
     """
 }
 
-process hog_rest{ //publishDir params.pickles_rhogs_folder
+process hog_rest{
+  publishDir params.pickles_temp
   input:
     val rhogsrest_tree_ready
   output:
-    path "pickle_rhogs/*.pickle"
-    // path "*.pickle"
-    //path "pickle_rhogs/*.pickle"
+    path "*.pickle"
     path "*.fa" , optional: true   // msa         if write True
     path "*.nwk" , optional: true  // gene trees  if write True
     val true
@@ -102,14 +97,13 @@ process hog_rest{ //publishDir params.pickles_rhogs_folder
 
 
 process collect_subhogs{
-  memory {4.GB}
   publishDir params.output_folder, mode: 'copy'
   input:
     val ready_hog_rest
-    val ready_hog_big     // path pickle_rhogs   // this is for depenedcy
-    path "pickle_rhogs/"  // "*.pickle"  // path "pickle_rhogs"   // this is the folder includes pickles_rhogs
+    val ready_hog_big
+    path "pickles_temp"   // this is the folder includes pickles_rhogs
     path "gene_id_dic_xml.pickle"
-    path "rhogs_all"
+    path "omamer_rhogs"
   output:
     path "output_hog.orthoxml"
     path "OrthologousGroupsFasta"
@@ -121,28 +115,25 @@ process collect_subhogs{
     """
 }
 
-
 workflow {
     proteomes = Channel.fromPath(params.proteomes,  type:'any' ,checkIfExists:true)
     proteome_folder = Channel.fromPath(params.proteome_folder)
     hogmap_folder = Channel.fromPath(params.hogmap_folder)
-//    rhogs_folder = Channel.fromPath(params.rhogs_folder)
 
     genetrees_folder = Channel.fromPath(params.genetrees_folder)
     hogmap_in = Channel.fromPath(params.hogmap_in)
 
-//    pickles_rhogs_folder =  Channel.fromPath(params.pickles_rhogs_folder)
-    omamerdb = Channel.fromPath(params.input_folder+"/omamerdb.h5")     // proteomes.view{"prot ${it}"}
+    pickles_temp =  Channel.fromPath(params.pickles_temp)
+    omamerdb = Channel.fromPath(params.input_folder+"/omamerdb.h5")
     proteomes_omamerdb = proteomes.combine(omamerdb)
     proteomes_omamerdb_inputhog = proteomes_omamerdb.combine(hogmap_in) // proteomes_omamerdb_inputhog.view{" rhogsbig ${it}"}
     (hogmap, ready_omamer_run)= omamer_run(proteomes_omamerdb_inputhog)
     ready_omamer_run_c = ready_omamer_run.collect()
 
-    (rhogs_folder, gene_id_dic_xml, ready_infer_roothogs) = infer_roothogs(ready_omamer_run_c, hogmap_folder, proteome_folder)
-    // rhogs_folder and "rhogs_all" are the same
+    (omamer_rhogs, gene_id_dic_xml, ready_infer_roothogs) = infer_roothogs(ready_omamer_run_c, hogmap_folder, proteome_folder)
     ready_infer_roothogs_c = ready_infer_roothogs.collect()
 
-    (rhogs_rest_list, rhogs_big_list, ready_batch_roothogs) = batch_roothogs(ready_infer_roothogs_c, rhogs_folder)
+    (rhogs_rest_list, rhogs_big_list, ready_batch_roothogs) = batch_roothogs(ready_infer_roothogs_c, omamer_rhogs)
     ready_batch_roothogs_c = ready_batch_roothogs.collect()
 
     species_tree = Channel.fromPath(params.species_tree)
@@ -155,16 +146,8 @@ workflow {
     rhogsrest_tree =  rhogsrest.combine(species_tree)
     rhogsrest_tree_ready = rhogsrest_tree.combine(ready_batch_roothogs_c)
     (pickle_rest_rhog,  msas_out_rest, genetrees_out_test, ready_hog_rest) = hog_rest(rhogsrest_tree_ready)
-    all_pickles = pickle_big_rhog.mix(pickle_rest_rhog).collect() // all_pickles.view() //     pickle_rhogs_folder = Channel.fromPath(params.output_folder+"/pickle_rhogs")
 
-    (orthoxml_file, OrthologousGroupsFasta, OrthologousGroups_tsv, rootHOGs_tsv)  = collect_subhogs(ready_hog_rest.collect(), ready_hog_big.collect(), all_pickles, gene_id_dic_xml, rhogs_folder)  // pickles_rhogs_folder
+    (orthoxml_file, OrthologousGroupsFasta, OrthologousGroups_tsv, rootHOGs_tsv)  = collect_subhogs(ready_hog_rest.collect(), ready_hog_big.collect(), pickles_temp, gene_id_dic_xml, omamer_rhogs)
     orthoxml_file.view{" output orthoxml file ${it}"}
 
 }
-
-// memory {12.GB * (2*task.attempt - 1)}
-//    time {24.hour}
-//    errorStrategy {
-//      task.exitStatus in [1,99,143,137,104,134,139,145,140] ? ‘retry’ : ‘terminate’
-//    }
-//    maxRetries 4