Merge pull request #146 from LANL-Bioinformatics/nf-gene-family-analysis

Nf gene family analysis

Author: yxu-lanl

workflows/Nextflow/configs/container.config

@@ -61,6 +61,12 @@ process {
     withLabel: 'phyl' {
         container = 'ghcr.io/lanl-bioinformatics/edge_snp_tree:0.6.0'
     }
+    withLabel: 'gfa' {
+        container = 'apwat/gene_family_analysis:pf2_0.3.3'
+    }
+    withLabel: 'vf' {
+        container = 'apwat/mvf:0.3'
+    }
     withLabel: 'report' {
         container = 'ghcr.io/lanl-bioinformatics/edge_report:0.5.0'
     }

workflows/Nextflow/main.nf

@@ -15,6 +15,8 @@ include {ANTISMASH} from './modules/runAntiSmash/runAntiSmash.nf'
 include {BINNING} from './modules/contigBinning/contigBinning.nf'
 include {REFERENCEBASEDANALYSIS} from './modules/referenceBasedAnalysis/refBasedAnalysis.nf'
 include {PHYLOGENETICANALYSIS} from './modules/SNPtree/SNPtree.nf'
+include {READSGENEFAMILYANALYSIS} from './modules/geneFamilyAnalysis/geneFamilyAnalysis.nf'
+include {CONTIGSGENEFAMILYANALYSIS} from './modules/geneFamilyAnalysis/geneFamilyAnalysis.nf'
 include {REPORT} from './modules/report/report.nf'
 
 workflow {
@@ -145,12 +147,16 @@ workflow {
     //Annotation and PhageFinder
     antismashInput = contigs
     annStats = channel.empty()
+    annGFF = channel.empty()
+    annFAA = channel.empty()
     if(params.modules.annotation) {
         ANNOTATION(baseSettings.plus(params.annotation), annContigs)
         annStats = ANNOTATION.out.annStats
+        annGFF = ANNOTATION.out.gff
+        annFAA = ANNOTATION.out.faa
 
         if(params.modules.phageFinder && (params.annotation.taxKingdom == null || !(params.annotation.taxKingdom.equalsIgnoreCase("viruses")))) {
-            PHAGEFINDER(baseSettings, ANNOTATION.out.gff, ANNOTATION.out.faa, ANNOTATION.out.fna)
+            PHAGEFINDER(baseSettings, annGFF, annFAA, ANNOTATION.out.fna)
         }
 
         antismashInput = ANNOTATION.out.gbk
@@ -171,6 +177,36 @@ workflow {
         PHYLOGENETICANALYSIS(baseSettings.plus(params.snpTree).plus(params.annotation), paired.ifEmpty(["${projectDir}/nf_assets/NO_FILE"]), unpaired.ifEmpty("${projectDir}/nf_assets/NO_FILE2"), contigs.ifEmpty("${projectDir}/nf_assets/NO_FILE3"))
     }
 
+    //gene family analysis
+        if(params.modules.readsGeneFamilyAnalysis) {
+        READSGENEFAMILYANALYSIS(baseSettings.plus(params.geneFamily), 
+            paired.ifEmpty(["${projectDir}/nf_assets/NO_FILE"]), 
+            unpaired.ifEmpty("${projectDir}/nf_assets/NO_FILE2"),
+        )
+    }
+
+    if(params.modules.contigsGeneFamilyAnalysis) {
+        geneFamilyFAA = channel.empty()
+        geneFamilyGFF = channel.empty()
+        if(params.geneFamily.inputFAA.endsWith("NO_FILE3")) {
+            geneFamilyFAA = annFAA.ifEmpty("${projectDir}/nf_assets/NO_FILE3")
+        }
+        else {
+            geneFamilyFAA = channel.fromPath(params.geneFamily.inputFAA)
+        }
+        if(params.geneFamily.inputGFF.endsWith("NO_FILE4")) {
+            geneFamilyGFF = annGFF.ifEmpty("${projectDir}/nf_assets/NO_FILE4")
+        }
+        else {
+            geneFamilyGFF = channel.fromPath(params.geneFamily.inputGFF)
+        }
+        CONTIGSGENEFAMILYANALYSIS(baseSettings.plus(params.geneFamily), 
+            geneFamilyFAA,
+            geneFamilyGFF,
+            contigs.ifEmpty("${projectDir}/nf_assets/NO_FILE5")
+        )
+    }
+
     //report generation
     REPORT(
         baseSettings, 

workflows/Nextflow/modules/geneFamilyAnalysis/Dockerfiles/MetaVF_Toolkit/Dockerfile

@@ -0,0 +1,38 @@
+# syntax=docker/dockerfile:1
+FROM continuumio/miniconda3:main AS build
+ENV version=0.1
+
+ENV container=docker
+
+RUN git clone --depth=1 https://github.com/aw-watson/MetaVF_toolkit.git
+
+# add conda channels
+RUN conda config --add channels conda-forge \
+    && conda config --add channels bioconda
+
+RUN conda init bash \
+    && . ~/.bashrc \
+    && conda env create --name MetaVF_toolkit -f /MetaVF_toolkit/env/MetaVF_toolkit_no_builds.yaml \
+    && conda activate MetaVF_toolkit 
+
+# #pack environment
+# RUN conda install -c conda-forge conda-pack
+
+# RUN conda-pack -n MetaVF_toolkit -o /tmp/env.tar && \
+#     mkdir /venv && cd /venv && tar xf /tmp/env.tar && \
+#     rm /tmp/env.tar
+
+# RUN /venv/bin/conda-unpack
+
+# FROM debian:latest AS runtime
+
+# COPY --from=build /venv /venv
+# COPY --from=build /MetaVF_toolkit /MetaVF_toolkit
+
+RUN gunzip /MetaVF_toolkit/databases/*.gz && chmod 755 /MetaVF_toolkit/metaVF.py
+
+ENV PATH=/venv/bin:/MetaVF_toolkit:$PATH
+RUN apt-get update && apt-get install procps -y && apt-get clean
+    
+SHELL ["/bin/bash", "-c"]
+CMD /bin/bash

workflows/Nextflow/modules/geneFamilyAnalysis/Dockerfiles/PathoFact2/Dockerfile

@@ -0,0 +1,56 @@
+# syntax=docker/dockerfile:1
+FROM continuumio/miniconda3:main AS build
+
+ENV container=docker
+
+# add conda channels
+RUN conda config --add channels conda-forge && \
+    conda config --add channels bioconda
+
+#TODO: evaluate if all environments here are required for running PathoFact2, given that our interest is in VF prediction
+RUN conda create -n rgi_env -c conda-forge -c bioconda rgi=6.0.3 -y 
+RUN conda create -n PathoFact_env -c conda-forge -c bioconda snakemake=7.25.0 Python=3.11.4 pandas=2.2.2 -y 
+RUN conda create -n deeparg_env  -c defaults  -c conda-forge -c bioconda python=2.7.18 diamond=0.9.24 cxx-compiler pip -y 
+RUN conda create -n amrfinder_plus_env -c conda-forge -c bioconda ncbi-amrfinderplus=3.12.8 -y 
+RUN conda create -n genomad -c conda-forge -c bioconda genomad=1.8.0 -y
+
+RUN conda init bash \
+    && . ~/.bashrc 
+
+#update databases for amrfinder_plus--the database update doesn't support a custom location, so we'll do it in the container
+#as opposed to putting the DBs elsewhere, which we do for other tools
+#TODO: evaluate impact of running this command during NF processes instead
+RUN /opt/conda/envs/amrfinder_plus_env/bin/amrfinder -u 
+
+#install DeepARG
+RUN /opt/conda/envs/deeparg_env/bin/pip install git+https://github.com/gaarangoa/deeparg.git
+
+#clone PF2
+RUN git clone https://gitlab.lcsb.uni.lu/ESB/PathoFact2.git --depth 1
+
+
+# create empty files to satisfy Snakemake rules, run Snakemake to generate needed conda environments
+# Without this step, running the Snakemake pipeline attempts to create those environments at runtime, and will create
+# errors due to running out of space in non-writeable containers
+RUN mkdir -p /abs/path/to/PathoFact2/DATABASES/HMM \
+    && mkdir -p /dry/run \
+    && mkdir /prefix \
+    && echo "#sample_name/folder,abs_path_to_contigs,input_type,extension(if folder)" > /abs/path/to/PathoFact2/list_of_samples.csv \
+    && echo "Test_contigs,/dry/run/contigs.fa,contigs" >> /abs/path/to/PathoFact2/list_of_samples.csv \
+    && echo "#end" >> /abs/path/to/PathoFact2/list_of_samples.csv \
+    && touch /dry/run/contigs.fa \
+    && touch /abs/path/to/PathoFact2/DATABASES/HMM/virulence_factors_CDD.hmm \
+    && touch /abs/path/to/PathoFact2/DATABASES/HMM/TOX_CDD.hmm \
+    && touch /abs/path/to/PathoFact2/DATABASES/cddid_all.tbl \
+    && conda run -n PathoFact_env snakemake -s /PathoFact2/Main.smk -c4 --configfile /PathoFact2/Config.yaml --use-conda --conda-frontend conda --conda-prefix /prefix  --conda-create-envs-only
+
+
+# deeparg download_data -o "<path_to_database_dir>/deeparg_db" #TODO: download DBs for DeepARG
+
+
+RUN apt-get update && apt-get install procps -y && apt-get clean
+ENV PATH=/PathoFact2:$PATH
+
+    
+SHELL ["/bin/bash", "-c"]
+CMD ["/bin/bash"]