add check for duplicate ONT read ids #592
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: compile and e.coli | |
| on: | |
| push: | |
| paths: | |
| - '**.sm' | |
| - '**.py' | |
| - '**.sh' | |
| - 'src/*' | |
| - '.github/workflows/*.yml' | |
| branches-ignore: | |
| - v2 | |
| pull_request: | |
| branches: [ master ] | |
| jobs: | |
| build_linux: | |
| name: Linux | |
| runs-on: ubuntu-latest | |
| steps: | |
| - name: Checkout | |
| uses: actions/checkout@v3 | |
| - name: Install Rust | |
| uses: actions-rs/toolchain@v1 | |
| with: | |
| toolchain: stable | |
| - name: Setup Python 3.11 | |
| uses: actions/setup-python@v2 | |
| with: | |
| python-version: 3.11 | |
| - name: Install Conda | |
| uses: conda-incubator/setup-miniconda@v3 | |
| with: | |
| auto-update-conda: true | |
| mamba-version: "*" | |
| channels: conda-forge,bioconda | |
| python-version: 3.11 | |
| - name: Install dependencies | |
| run: | | |
| mamba install -c conda-forge -c bioconda 'python=3.11' 'pulp<=2.7.0' 'snakemake-minimal>7,<8' GraphAligner mashmap winnowmap | |
| sudo apt-get install libcurl4-openssl-dev | |
| pip install networkx | |
| echo "" | |
| echo "CONDA LIST" | |
| echo "" | |
| conda list | |
| echo "" | |
| echo "PYTHON TEST" | |
| echo " "`command -v python` | |
| echo " "`python --version` | |
| echo > x.py "from importlib import metadata" | |
| echo >> x.py "for dist in metadata.distributions():" | |
| echo >> x.py " print(dist._path)" | |
| python x.py | |
| echo > x.py "from importlib.metadata import version" | |
| echo >> x.py "version('networkx')" | |
| python x.py | |
| echo > x.py "import networkx as nx" | |
| echo >> x.py "G = nx.Graph()" | |
| echo >> x.py "print(G)" | |
| python x.py | |
| - name: Compile | |
| run: | | |
| cd src | |
| make -j 2 | |
| - name: Fetch Data | |
| run: | | |
| curl -L https://obj.umiacs.umd.edu/sergek/shared/ecoli_hifi_subset24x.fastq.gz -o hifi.fastq.gz | |
| curl -L https://obj.umiacs.umd.edu/sergek/shared/ecoli_ont_subset50x.fastq.gz -o ont.fastq.gz | |
| curl -L https://obj.umiacs.umd.edu/sergek/shared/ecoli-test-cache.tar -o cache.tar | |
| tar -xf cache.tar | |
| - name: Assemble | |
| run: | | |
| reportLogs() { | |
| for fn in `find . -name \*err -print` ; do | |
| echo "" | |
| echo "==BEGIN $fn" | |
| cat $fn | |
| echo "==END $fn" | |
| done | |
| exit 1 | |
| } | |
| export PATH=${PATH}:$CONDA/bin | |
| rm -f mbg.sh | |
| rm -f ga.sh | |
| rm -f lib/verkko/bin/utgcns | |
| echo > mbg.sh '#!/bin/sh' | |
| echo >> mbg.sh 'xz -dc ../../cache/hifi-resolved.gfa.xz > hifi-resolved.gfa' | |
| echo >> mbg.sh 'cp ../../cache/hifi_nodecov.csv hifi_nodecov.csv' | |
| echo >> mbg.sh 'xz -dc ../../cache/paths.gaf.xz > paths.gaf' | |
| echo > ga.sh '#!/bin/sh' | |
| echo >> ga.sh 'if [ ! -e "graph.index" ] ; then' | |
| echo >> ga.sh ' touch graph.index' | |
| echo >> ga.sh ' touch diploid.index' | |
| echo >> ga.sh ' touch empty.gaf' | |
| echo >> ga.sh 'else' | |
| echo >> ga.sh ' xz -dc ../../cache/aligned001.WORKING.gaf.xz > aligned001.WORKING.gaf' | |
| echo >> ga.sh 'fi' | |
| echo > lib/verkko/bin/utgcns '#!/bin/sh' | |
| echo >> lib/verkko/bin/utgcns 'if [ ! -e part001.fasta ]; then' | |
| echo >> lib/verkko/bin/utgcns 'cp ../../cache/part001.fasta packages/part001.fasta.WORKING' | |
| echo >> lib/verkko/bin/utgcns 'fi' | |
| echo >> lib/verkko/bin/utgcns '' | |
| echo >> lib/verkko/bin/utgcns 'if [ ! -e part002.fasta ]; then' | |
| echo >> lib/verkko/bin/utgcns 'cp ../../cache/part002.fasta packages/part002.fasta.WORKING' | |
| echo >> lib/verkko/bin/utgcns 'fi' | |
| chmod 755 mbg.sh | |
| chmod 755 ga.sh | |
| chmod 755 lib/verkko/bin/utgcns | |
| ./bin/verkko -d asm --mbg $PWD/mbg.sh --graphaligner $PWD/ga.sh --hifi ./hifi.fastq.gz --nano ./ont.fastq.gz | |
| if [ ! -s asm/assembly.fasta ]; then | |
| reportLogs | |
| fi |