Repository to keep track of miscellaneous SARS-CoV-2 analysis scripts
snp_analysis_plot.py
Quick script that generates a heatmap figure from a series of SnpEff annotated freebayes generated VCF files (ideally SARS-CoV-2).
Requires pandas pyvcf and searborn:
conda create -n snp_analysis pandas pyvcf seaborn
conda activate snp_analysis
python snp_analysis_plot.py -n test_snp_plot test/21-B1428411R.ann.vcf test/21-B1428413R.ann.vcf test/21-B1428417R.ann.vcf test/21-B1428418R.ann.vcf
>>> Writing figure to test_snp_plot_passaging_all.png
Quick script to translate and provide quick human readable breakdown of mutations from ivar variants. Hopefully brings some of the convenience of the nextclade summaries using the more robust ivar inferences instead of relying on minimap2 alignments.
Very loosely based on parts of type_variants.
Requires gffutils and pandas to work:
conda create -n summarise_ivar gffutils pandas
conda activate summarise_ivar
python summarise_ivar.py --input test/test1_ivar_variants.tsv test/test2_ivar_variants.tsv --output_type summary
>>> test/test1_ivar_variants.tsv: ['aa:N:R203K', 'aa:N:G204R', 'snp:N:G28882A', 'aa:S:D614G', 'snp:non-coding:C241T', 'aa:orf1ab:P4715L', 'snp:orf1ab:C3037T', 'snp:orf1ab:T7288C', 'snp:orf1ab:A14199G']
>>> test/test2_ivar_variants.tsv: ['aa:N:R203K', 'aa:N:G204R', 'snp:non-coding:C241T', 'snp:orf1ab:C3037T', 'snp:orf1ab:T7288C']
Generate pangolin and nextclade lineage assignments before collating results into a single table containing version metadata for tools and data dependencies
Requires nextclade and pangolin to work:
conda env create -f https://raw.githubusercontent.com/cov-lineages/pangolin/master/environment.yml
conda activate pangolin
pip install git+https://github.com/cov-lineages/pangolin
pip install git+https://github.com/cov-lineages/pangoLEARN.git --upgrade
pip install git+https://github.com/cov-lineages/lineages.git --upgrade
conda install -y nodejs
npm install --global @neherlab/nextclade
python assign_lineages.py --input_genomes test/genomes.fa --output lineages.tsv
Extract sequences and metadata from nextstrain ingested GISAID dumps based on a pandas query of the metadata e.g. all P.1 sequences.
Requires pandas and biopython to work:
conda create -n extract_sequences biopython pandas
conda activate extract_sequences
python extract_seqs.py --nextmeta metadata_2021-01-08_18-19.tsv --nextfasta sequences_2021-01-08_08-46.fasta --query "pangolin_lineage=='B.1.1.28'" --output_prefix filt_test
Quick tool that uses the upset_plotly library (a plotly version of upsetplot) to parse the outputs of a variant checking tool and generate an interactive upsetplot showing the co-occurrence of specific variants.
Supported inputs:
- Output csv from type_variants (must be run with --append-genotypes)
- Output from ncov-watch.py.
python -m pip install git+git://github.com/fmaguire/upset_plotly
To create a plot for an output from type_variants:
python variant_intersections.py -i variant_check.tsv -t type_variants -o example.html
Goes through a nextmeta formatted metadata tsv and for any Canadian postcode
entered into the location field (not already in ncov lat_long.tsv file)
uses pgeocode to get the correct latitude and longitude.
This geolocation information is then appended to the ncov lat_long.tsv file
and output wherever the user specifies.
Required pandas and pgeocode
conda create -n pgeocode pandas pip
pytohn -m pip install pgeocode
python get_lat_long_for_postcode.py --metadata metadata_2021-01-08_18-19.tsv --lat_longs lat_longs.tsv --output custom_lat_longs.tsv
Takes a nextclade output tsv and extracts the mutations/deletions for all (default) or a specific gene
Requires pandas
python collect_mutations.py --metadata metadata.tsv --nextclade nextclade.csv --gene S --output metadata_with_mutations.tsv