We're going to work with data from a human cell line posted here: https://storage.googleapis.com/bfx_workshop_tmp/Exome_Tumor.tar

• Make a directory called "week03", and download the tarball to your computer using the command line (wget or curl).

• Use tar -xvf to extract the directory from the tar file, then cd into the directory and look around with ls. We're not going to use all of this data in this week's homework. Let's focus on the contents of Exome_Tumor.tar. Untar it, then unzip the fastq files.

• Look at the first three records (not first three lines!) of each fastq file. Take a close look at the read names and how they match up across files.

On your laptop:

docker pull ubuntu

What is that doing? It's going to https://hub.docker.com/_/ubuntu and pulling down the image with the "latest" tag

docker run ubuntu

• Perform single-sample germline variant calling with GATK HaplotypeCaller and the GATK GVCF workflow on exome data
• Perform joint genotype calling on exome data, including additional exomes from 1000 Genomes Project
• Manipulate and Filter VCFs to remove artifacts and identify variants of interest

In this module we will use the GATK HaplotypeCaller to call germline variants from "normal" bams. For a more in-depth look, see this excellent GATK tutorial, provided by the Broad Institute.

Let's start by looking at some outputs of long read sequencing from the Oxford Nanopore (ONT) platform. These are sequences from the K562 cell line, prepared with the ONT cDNA sequencing kit (poly-A selected). Off the machines, the data will consist of a FAST5 or POD5 file, which are a compressed representation of the raw signal. These are subsequently run through a basecalling algorithm (such as Dorado) to generate FASTQ files.

The choice of basecalling algorithm and parameters goes pretty deep, so we'll assume that reasonable choices have been made. For simplicity, we've also subset the data to include just small portions of the genome, including a few genes of interest.

Go ahead and pull down this fastq file:

wget https://storage.googleapis.com/bfx_workshop_tmp/k562_ont_raw.fastq.gz

Start by gathering some data. Navigate to a somatic folder and pull down a set of input data from (human build38) from this location:

wget https://storage.googleapis.com/bfx_workshop_tmp/inputs.tar.gz

There are three inputs you need to run a workflow:

1. A .cwl file that contains the steps to be run
2. A .yaml file that gives the inputs to that CWL
3. A config file that tells cromwell about it's environment, how to submit jobs to the cluster, and where to stick the results

Let's start with #3 - the config file. I've made this easy for you. Create a directory where you want to run things, then inside of it, run the following command:

/storage1/fs1/timley/Active/aml_ppg/src/utilities/create_cromwell_config -o cromwell.config -l logs -d output -q timley -G compute-timley```

• Join the mgibio Slack. Ask any user to invite you using your wustl address, or email [email protected]. Excellent place to post questions about anything and get answers. Useful channels include #bfx-workshop, #analysis-workflows, #cancergenomics, and #docker
• Get compute1 access set up. This requires a ticket to the RIS Servicedesk requesting to be added to the appropriate compute and storage groups
• VPN access Connect to msvpn.wusm.wustl.edu through Cisco AnyConnect. Use WUSTL key log in and submit request at https://it.wustl.edu/items/connect/
• Set up compute1 config files (Need a link for this - env variables, etc)
• Sign up for the bfx_workshop get on the [email list](https://outlook.office365.com/owa/[email protected]/groupsubscription.ashx?action=join&source=MSExchange/LokiServer&guid=

• Add RNAseq sanity/QC checks https://github.com/genome/analysis-workflows/issues/904
• Add input type fastqs to RNA seq https://github.com/genome/analysis-workflows/issues/932

Open up a terminal (Terminal/iTerm on Mac, putty or WSL on Windows) and SSH into the cluster, replacing USERNAME with your WUSTL key.

ssh [email protected]