Clinical Reporting in R
This is a prototype implementation of a clinical reporting pipeline in R.
Currently, it creates a genetic report of somatic mutations from a vcf file annotated via Ensembl Variant Effect Predictor.
Note that this is an experimental branch. Currently, we need to build a docker image from this repository, but we can't run the
reporting.R script from within the container, because it requires access to the REST service above, which will run on localhost port 5000. I haven't figured out yet how to make this accessible from the container, so we run the annotation from the container but create the final report using the plain reporting.R script.
We assume that we want to create a report for a vcf file
my.vcf residing in
$HOME, and that you have R and Docker installed. First, follow instructions to install and run the Biograph REST API. Then clone this repository and
cd into it.
First, we need to get and run the REST service via docker:
1. git clone https://github.com/mrdivine/clinicalReporting_DB_RESTAPI.git 2. cd clinicalReporting_DB_RESTAPI 3. docker-compose up
Now we clone this repository and checkout the single_script branch:
4. git clone -b single_script https://github.com/PersonalizedOncology/ClinicalReportR.git 5. cd ClinicalReportR 6. docker build -t personalizedoncology/clinicalreportr:latest . 7. export CLINICALREPORTR=`pwd`
Now we download some data for required for ensemble-vep:
8. wget http://www.broadinstitute.org/~konradk/loftee/human_ancestor.fa.rz 9. wget http://www.broadinstitute.org/~konradk/loftee/human_ancestor.fa.rz.fai 10. wget https://github.com/Ensembl/VEP_plugins/blob/release/90/LoFtool_scores.txt 11. docker run -t -i -v $CLINICALREPORTR:/home/vep/.vep personalizedoncology/clinicalreportr:latest perl /home/vep/src/ensembl-vep/INSTALL.pl perl INSTALL.pl -a cf -s homo_sapiens -y GRCh38
This will download cache files into
$CLINICALREPORTR, e.g. the directory you cloned this repository into.
11. docker run -t -i -v $CLINICALREPORTR:/data personalizedoncology/clinicalreportr:latest vep -i /data/strelka.passed.missense.somatic.snvs_test.vcf -o my_annotated.vcf 12. Rscript reporting.R -f my_annotated.vcf
You should now have a file