Update VEP to release 101 (August 2020). (#11)

* Update VEP to release 101 (August 2020).

* Small updates from 91->101; code comment path change.

batch/vep/Dockerfile

@@ -17,7 +17,7 @@
 #
 # Example:
 #
-# docker build . --build-arg ENSEMBL_RELEASE=91 --tag vep_91
+# docker build . --build-arg ENSEMBL_RELEASE=101 --tag vep_101
 #
 # To run vep through containers created by this file, the VEP cache has to be
 # downloaded separately and made available through command line arguments.
@@ -27,13 +27,15 @@
 # retry logic.
 FROM gcr.io/cloud-genomics-pipelines/io
 
-ARG ENSEMBL_RELEASE=91
+ARG ENSEMBL_RELEASE=101
 ARG VEP_BASE=/opt/variant_effect_predictor
 
 RUN apt-get -y update && apt-get install -y procps\
     build-essential \
     git \
     libarchive-zip-perl \
+    libbz2-dev \
+    liblzma-dev \
     libdbd-mysql-perl \
     libdbi-perl \
     libfile-copy-recursive-perl \

batch/vep/README.md

@@ -27,17 +27,17 @@ Inside this directory, run:
 
 This will download the source from
 [VEP GitHub repo](https://github.com/Ensembl/ensembl-vep) and build VEP from
-that source. By default, it uses version 91 of VEP. This can be changed by
+that source. By default, it uses version 101 of VEP. This can be changed by
 `ENSEMBL_RELEASE` build argument, e.g.,
 
 `docker build . -t [IMAGE_TAG] --build-arg ENSEMBL_RELEASE=90`
 
 Let's say we want to push this image to the
 [Container Registry](https://cloud.google.com/container-registry/) of
 `my-project` on Google Cloud, so we can pick `[IMAGE_TAG]` as
-`gcr.io/my-project/vep:91`. Then push this image by:
+`gcr.io/my-project/vep:101`. Then push this image by:
 
-`gcloud docker -- push gcr.io/my-project/vep:91`
+`gcloud docker -- push gcr.io/my-project/vep:101`
 
 **TODO**: Add `cloudbuild.yaml` files for both easy push and integration test.
 
@@ -48,7 +48,7 @@ download and integrate different pieces of the VEP database or cache files.
 Then from within that directory run the
 [`build_vep_cache.sh`](build_vep_cache.sh) script. By default this script
 creates the database for human (homo_sapiens), referenec sequence `GRCh38`,
-and release 91 of VEP. These values can be overwritten by the following
+and release 101 of VEP. These values can be overwritten by the following
 environment variables (note you should use the same VEP release
 that you used for creating VEP docker image above):
 
@@ -74,7 +74,7 @@ gcloud alpha genomics pipelines run \
   --inputs VCF_INFO_FILED=CSQ_RERUN
 ```
 
-Note the `vep_cache_homo_sapiens_GRCh38_91.tar.gz` file that is referenced in
+Note the `vep_cache_homo_sapiens_GRCh38_101.tar.gz` file that is referenced in
 the sample `yaml` file, is the output file that you get from the above database
 creation step.
 

batch/vep/build_vep_cache.sh

@@ -26,14 +26,14 @@
 # Capital letter variables refer to environment variables that can be set from
 # outside. Internal variables have small letters. All environment variables
 # have a default value as well to set up cache for homo_sapiens with reference
-# GRCh38 and release 91 of VEP.
+# GRCh38 and release 101 of VEP.
 #
 # More details on cache files can be found here:
 # https://ensembl.org/info/docs/tools/vep/script/vep_cache.html
 
 set -euo pipefail
 
-readonly release="${ENSEMBL_RELEASE:-91}"
+readonly release="${ENSEMBL_RELEASE:-101}"
 readonly species="${VEP_SPECIES:-homo_sapiens}"
 readonly assembly="${GENOME_ASSEMBLY:-GRCh38}"
 readonly work_dir="vep_cache"
@@ -81,7 +81,12 @@ else
   curl -O "${remote_fasta}.gzi"
 fi
 
-readonly remote_cache="${ftp_base}/variation/VEP/${cache_file}"
+# The path naming convention changed from "VEP" to "vep" after build 95.
+if (( release <= 95 )); then
+  readonly remote_cache="${ftp_base}/variation/VEP/${cache_file}"
+else
+  readonly remote_cache="${ftp_base}/variation/vep/${cache_file}"
+fi
 echo "Downloading ${remote_cache} ..."
 curl -O "${remote_cache}"
 echo "Decompressing cache files ..."