variant_QC

Sequencing errors and artefacts are important confounding factors for detection of genetic variants in clinical settings (e.g for diagnostic classification or prognosic insights) based on deep next-generation sequencing. Targeted deep sequencing is usually done by amplicon or hybridization-capture protocol that are also potential sources of errors. The current case study excluded non-sequencer related errors by evaluating the perfomance of two sequencers on the same sample prepared with the same laboratory protocols.

Visualising the results

The results are visualised in Jupyter notebooks and single html page IGV viewers generated as part of the pipeline.

• Jupyter notebooks can be visualised on Github or locally
• IGV html can only be visualised locally (downloading just the html file will also work, e.g. data/processed/igv_viewer.html)

Reproducing the results

To reproduce full results run the steps listed below. Alternatively, Jupyter notebooks can be rerun on pre-computed files (if all steps were run).

Step 1 - Build docker image

Clone the repository and build docker image.

git clone https://github.com/ksanao/variant_QC.git  
cd variant_QC  
docker build -t variant_qc .  
cd ..  

Step 2 - Start docker container and Jupyter Lab

To run docker image and start the container run the command below. The repository directory will be mounted in Docker container.

variant_QC/start_docker.sh run  

Jupyter lab will be available at http://127.0.0.1:8888/lab (provde the token printed out in the container). 
To print again running labs and tokens run the following command inside container:
jupyter notebook list

Step 3 - Run the pipeline comparing sequencers

Inside container run the following commands

source .bashrc   
compare.sh src/config

Step 4 - Examine the results of Step 3 in Jupyter notebook

File notebooks/01_compare_sequencers.ipynb

Step 5 - Run variant calling pipeline

Inside container run the following commands

source .bashrc  
call_variants.sh src/config SG001_1.bam

Step 6 - Examine the results of Step 5 in Jupyter notebook

File notebooks/02_variant_calls.ipynb