Refactor vrs_map to use mavehgvs

.github/workflows/checks.yaml

@@ -6,7 +6,7 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       matrix:
-        python-version: ["3.10", "3.11", "3.12"]
+        python-version: ["3.11", "3.12"]
     steps:
       - uses: actions/checkout@v3
 

.gitignore

@@ -161,6 +161,7 @@ cython_debug/
 .idea/
 
 *.pickle
+.vscode
 
 # mapping data/output
 notebooks/analysis/analysis_files

.python-version

@@ -0,0 +1 @@
+3.11.4

Dockerfile

@@ -0,0 +1,51 @@
+FROM --platform=x86_64 python:3.11
+
+RUN apt update
+# Install tools necessary used to install samtools and htslib so we can configure fasta files for genomic assembly.
+RUN apt-get clean && apt-get update && apt-get install -y \
+    postgresql-client \
+	build-essential \
+	curl \
+	git \
+	libbz2-dev \
+	libcurl4-openssl-dev \
+	libgsl0-dev \
+	liblzma-dev \
+	libncurses5-dev \
+	libperl-dev \
+	libssl-dev \
+	zlib1g-dev \
+    && rm -rf /var/lib/apt/lists/*
+
+# download and install blat executable
+WORKDIR /usr/bin
+RUN wget http://hgdownload.cse.ucsc.edu/admin/exe/linux.x86_64/blat/blat
+RUN chmod +x blat
+
+# set dcd_mapping resources directory and download reference file
+WORKDIR /home/.local/share/dcd_mapping
+ENV DCD_MAPPING_RESOURCES_DIR=/home/.local/share/dcd_mapping
+RUN curl -LJO https://hgdownload.cse.ucsc.edu/goldenpath/hg38/bigZips/hg38.2bit
+
+# Install samtools and htslib.
+ARG htsversion=1.19
+RUN curl -L https://github.com/samtools/htslib/releases/download/${htsversion}/htslib-${htsversion}.tar.bz2 | tar xj && \
+    (cd htslib-${htsversion} && ./configure --enable-plugins --with-plugin-path='$(libexecdir)/htslib:/usr/libexec/htslib' && make install) && \
+    ldconfig && \
+    curl -L https://github.com/samtools/samtools/releases/download/${htsversion}/samtools-${htsversion}.tar.bz2 | tar xj && \
+    (cd samtools-${htsversion} && ./configure --with-htslib=system && make install) && \
+    curl -L https://github.com/samtools/bcftools/releases/download/${htsversion}/bcftools-${htsversion}.tar.bz2 | tar xj && \
+    (cd bcftools-${htsversion} && ./configure --enable-libgsl --enable-perl-filters --with-htslib=system && make install)
+
+RUN mkdir /usr/src/app
+WORKDIR /usr/src/app
+COPY . .
+
+RUN pip install -e '.[dev,tests]'
+# use polars-lts-cpu to avoid issues with x86 emulation on arm machine
+RUN pip install -U polars-lts-cpu
+# install gene normalizer with pg dependencies. TODO: can the pg dependencies be specified in pyproject.toml?
+#RUN pip install 'gene-normalizer[pg]'
+ENV PYTHONUNBUFFERED 1
+
+ENV PYTHONPATH "${PYTHONPATH}:/usr/src/app/src"

README.md

@@ -44,6 +44,14 @@ Use `dcd-map --help` to see other available options.
 
 Notebooks for manuscript data analysis and figure generation are provided within `notebooks/analysis`. See [`notebooks/analysis/README.md`](notebooks/analysis/README.md) for more information.
 
+Following installation instructions for [CoolSeqTool](https://coolseqtool.readthedocs.io/latest/install.html) and [Gene Normalizer](https://gene-normalizer.readthedocs.io/latest/install.html) should take care of the external data dependencies.
+
+Note that Gene Normalizer's `pg` dependency group must be installed to make use of the PostgreSQL-based backend:
+
+```shell
+python3 -m pip install 'gene-normalizer[pg]'
+```
+
 ## Development
 
 Clone the repo

docker-compose-dev.yml

@@ -0,0 +1,39 @@
+version: "3"
+
+services:
+  app:
+    build: .
+    command: bash -c "tail -f /dev/null"
+    depends_on:
+      - db
+      - seqrepo
+    env_file:
+      - settings/.env.dev
+    environment:
+      DB_HOST: db
+      DB_PORT: 5432
+    ports:
+      - "8002:8000"
+    volumes:
+      - .:/usr/src/app
+      - vrs-mapping-seqrepo-dev:/usr/local/share/seqrepo
+
+  db:
+    image: postgres:14
+    env_file:
+      - settings/.env.dev
+    ports:
+      - "5434:5432"
+    expose:
+      - 5432
+    volumes:
+      - vrs-mapping-data-dev:/var/lib/postgresql/data
+
+  seqrepo:
+    image: biocommons/seqrepo:2021-01-29
+    volumes:
+      - vrs-mapping-seqrepo-dev:/usr/local/share/seqrepo
+
+volumes:
+  vrs-mapping-data-dev:
+  vrs-mapping-seqrepo-dev:

notebooks/analysis/mave_mapping_fig_3b.R

@@ -37,7 +37,7 @@ df <- data.frame('Experiment Cellular Context' = names, 'value' = context_counts
 
 ggplot(df, aes(x = factor(Experiment.Cellular.Context, levels = c('Human', 'Yeast', 'Bacteria', 'Mouse', 'Bacteriophage', 'N/A')), y = value, fill = rownames(df))) +
   geom_bar(stat = 'identity', fill = c("#F8766D","#B79F00","#90ee90","#00BFC4","#619CFF","#F564E3")) +
-  geom_text(aes(label = value), vjust = ifelse(df$value != 92, -1, 3), size = 10, colour = ifelse(df$value == 97, 'white', 'black')) +
+  geom_text(aes(label = value), vjust = ifelse(df$value != 86, -1, 3), size = 10, colour = ifelse(df$value == 97, 'white', 'black')) + 
   xlab('MAVE Experiment Cellular Context') +
   ylab('Number of Experiments') +
   scale_y_continuous(expand = c(0, 0)) +