Merge branch 'development' into no_upload_master

cidgoh · Mar 8, 2023 · 2d7b37c · 2d7b37c
2 parents 1f7fe5f + 967d8bc
commit 2d7b37c
Show file tree

Hide file tree

Showing 1,402 changed files with 222,514 additions and 158,438 deletions.
diff --git a/data_parser.py b/data_parser.py
@@ -111,46 +111,58 @@ def parse_gvf_dir(dir_):
                 pos = row["#start"]
                 if pos not in ret[strain]["mutations"]:
                     ret[strain]["mutations"][pos] = []
-                    mutation_types = row["#type"].split(",")
-                    num_of_mutations = len(mutation_types)
-                    for i in range(num_of_mutations):
-                        mutation_dict = {
-                            "ref": attrs["Reference_seq"],
-                            "alt": attrs["Variant_seq"].split(",")[i],
-                            "gene": attrs["vcf_gene"],
-                            "ao": float(attrs["ao"].split(",")[i]),
-                            "dp": float(attrs["dp"]),
-                            "multi_aa_name": attrs["multi_aa_name"],
-                            "clade_defining":
-                                attrs["clade_defining"] == "True",
-                            "hidden_cell": False,
-                            "mutation_name": attrs["Name"],
-                            "functions": {}
-                        }
-                        alt_freq = mutation_dict["ao"]/mutation_dict["dp"]
-                        mutation_dict["alt_freq"] = str(round(alt_freq, 4))
-                        type = mutation_types[i]
-                        if type == "ins":
-                            mutation_dict["mutation_type"] = "insertion"
-                        elif type == "del":
-                            mutation_dict["mutation_type"] = "deletion"
-                        else:
-                            mutation_dict["mutation_type"] = "snp"
-                        ret[strain]["mutations"][pos].append(mutation_dict)
 
+                mutation_name = attrs["Name"]
+                alt = attrs["Variant_seq"]
+
+                mutation_dict = {}
+                for existing_dict in ret[strain]["mutations"][pos]:
+                    cond1 = existing_dict["mutation_name"] == mutation_name
+                    cond2 = existing_dict["alt"] == alt
+                    if cond1 and cond2:
+                        mutation_dict = existing_dict
+                        break
+
+                if not mutation_dict:
+                    mutation_dict = {
+                        "ref": attrs["Reference_seq"],
+                        "alt": alt,
+                        "gene": attrs["vcf_gene"],
+                        "ao": float(attrs["ao"]),
+                        "dp": float(attrs["dp"]),
+                        "multi_aa_name": attrs["multi_aa_name"],
+                        "clade_defining":
+                            attrs["clade_defining"] == "True",
+                        "hidden_cell": False,
+                        "mutation_name": mutation_name,
+                        "functions": {}
+                    }
+
+                    alt_freq = mutation_dict["ao"]/mutation_dict["dp"]
+                    mutation_dict["alt_freq"] = str(round(alt_freq, 4))
+
+                    mutation_type = row["#type"]
+                    if mutation_type == "ins":
+                        mutation_dict["mutation_type"] = "insertion"
+                    elif mutation_type == "del":
+                        mutation_dict["mutation_type"] = "deletion"
+                    else:
+                        mutation_dict["mutation_type"] = mutation_type
+
+                    ret[strain]["mutations"][pos].append(mutation_dict)
+
+                fn_dict = mutation_dict["functions"]
                 fn_category = attrs["function_category"].strip('"')
+                if not fn_category:
+                    continue
+                if fn_category not in fn_dict:
+                    fn_dict[fn_category] = {}
+
                 fn_desc = attrs["function_description"].strip('"')
                 fn_source = attrs["source"].strip('"')
                 fn_citation = attrs["citation"].strip('"')
-                fn_dict = {}
-                if fn_category:
-                    if fn_category not in fn_dict:
-                        fn_dict[fn_category] = {}
-                    fn_dict[fn_category][fn_desc] = \
-                        {"source": fn_source, "citation": fn_citation}
-                for i in range(len(ret[strain]["mutations"][pos])):
-                    parsed_mutation = ret[strain]["mutations"][pos]
-                    parsed_mutation[i]["functions"].update(fn_dict)
+                fn_dict[fn_category][fn_desc] = \
+                    {"source": fn_source, "citation": fn_citation}
 
     return ret
 

diff --git a/docker-compose.yml b/docker-compose.yml
@@ -0,0 +1,13 @@
+version: '3'
+
+services:
+  app:
+    tty: true
+    build:
+      context: .
+      dockerfile: ./dockerConf/Dockerfile
+    image: covid_mvp_app
+    ports:
+      - "8050:8050"
+    volumes:
+      - .:/app
diff --git a/dockerConf/Dockerfile b/dockerConf/Dockerfile
@@ -0,0 +1,19 @@
+FROM continuumio/miniconda3
+
+WORKDIR /app
+
+COPY ./dockerConf/environment.yml /environment.yml
+COPY ./dockerConf/start.sh /start.sh
+
+RUN apt-get update -y \
+    && apt-get install default-jre -y \
+    && wget -qO- https://get.nextflow.io | bash \
+    && chmod +x nextflow \
+    && mv /app/nextflow /usr/local/bin \
+    && conda env create -f /environment.yml \
+    && sed -i 's/\r$//g' /start.sh \
+    && chmod +x /start.sh
+
+# https://pythonspeed.com/articles/activate-conda-dockerfile/
+ENTRYPOINT ["conda", "run", "--no-capture-output", "-n", "COVID-MVP", "/start.sh"]
+EXPOSE 8050
diff --git a/dockerConf/app/Dockerfile b/dockerConf/app/Dockerfile
diff --git a/dockerConf/app/condaEnviroment.yml b/dockerConf/app/condaEnviroment.yml
diff --git a/dockerConf/environment.yml b/dockerConf/environment.yml
@@ -0,0 +1,55 @@
+name: COVID-MVP
+channels:
+  - conda-forge
+  - bioconda
+  - defaults
+dependencies:
+  - _libgcc_mutex=0.1=conda_forge
+  - _openmp_mutex=4.5=2_gnu
+  - bzip2=1.0.8=h7f98852_4
+  - ca-certificates=2023.01.10=h06a4308_0
+  - ld_impl_linux-64=2.40=h41732ed_0
+  - libffi=3.4.2=h7f98852_5
+  - libgcc-ng=12.2.0=h65d4601_19
+  - libgomp=12.2.0=h65d4601_19
+  - libnsl=2.0.0=h7f98852_0
+  - libsqlite=3.40.0=h753d276_0
+  - libuuid=2.32.1=h7f98852_1000
+  - libzlib=1.2.13=h166bdaf_4
+  - ncurses=6.4=h6a678d5_0
+  - openssl=3.0.8=h0b41bf4_0
+  - pip=23.0.1=pyhd8ed1ab_0
+  - python=3.9.16=h2782a2a_0_cpython
+  - readline=8.2=h5eee18b_0
+  - setuptools=67.5.1=pyhd8ed1ab_0
+  - tk=8.6.12=h27826a3_0
+  - tzdata=2022g=h191b570_0
+  - wheel=0.38.4=pyhd8ed1ab_0
+  - xz=5.2.10=h5eee18b_1
+  - pip:
+    - brotli==1.0.9
+    - click==8.1.3
+    - dash==1.20.0
+    - dash-bootstrap-components==0.11.3
+    - dash-core-components==1.16.0
+    - dash-html-components==1.1.3
+    - dash-renderer==1.9.1
+    - dash-table==4.11.3
+    - flask==2.1.3
+    - flask-caching==1.10.1
+    - flask-compress==1.13
+    - future==0.18.3
+    - gunicorn==20.1.0
+    - importlib-metadata==6.0.0
+    - itsdangerous==2.1.2
+    - jinja2==3.1.2
+    - markupsafe==2.1.2
+    - numpy==1.24.2
+    - pandas==1.2.0
+    - plotly==5.13.1
+    - python-dateutil==2.8.2
+    - pytz==2022.7.1
+    - six==1.16.0
+    - tenacity==8.2.2
+    - werkzeug==2.0.2
+    - zipp==3.15.0
diff --git a/dockerConf/nginx/Dockerfile b/dockerConf/nginx/Dockerfile
diff --git a/dockerConf/nginx/nginx.conf b/dockerConf/nginx/nginx.conf
diff --git a/dockerConf/app/start.sh → dockerConf/start.sh b/dockerConf/app/start.sh → dockerConf/start.sh
diff --git a/local.yml b/local.yml
diff --git a/nf-ncov-voc b/nf-ncov-voc
diff --git a/production.yml b/production.yml