EXCAVATOR2_free

This repository hosts a copy of the EXCAVATOR2 copy number caller hosted on sourceforge at https://sourceforge.net/projects/excavator2tool/ which has been modified in a way that allows it to be run outside of the directory in which the tool is stored.
If you use EXCAVATOR2 in your work, please cite the original authors and publication:

D'Aurizio, R., Pippucci, T., Tattini, L., Giusti, B., Pellegrini, M. and Magi, A., 2016. Enhanced copy number variants detection from whole-exome sequencing data using EXCAVATOR2. Nucleic acids research, 44(20), pp.e154-e154.

**Please refer to the original documentation, hosted in EXCAVATOR2_free/docs/ for any questions regarding the functions and parameters used in EXCAVATOR2. **

Prerequisites

R package Hmisc
Perl
R
samtools
bedtools
ucscsuite
GCA_000001405.15_GRCh38.bw and/or ucsc.hg19.bw bw file saved in EXCAVATOR2_free/data/

Installation and preparation

git clone https://github.com/mjakobs/EXCAVATOR2_free.git
cd EXCAVATOR2_free/data/
mkdir targets
cd targets/
mkdir hg19
mkdir hg38

Preparation of the target file

Must be done in the EXCAVATOR2_free/ directory.
Make sure perl, R, samtools, bedtools, and ucscsuite are loaded in your environment.

perl TargetPerla.pl SourceTarget_GRCh38.txt /path/to/your/target.bed name-of-your-target_w50000 50000 hg38

Window size can be adjusted by changing 50000 to another value, and reference genome can be adjusted by changing hg38.
This step will create the reference files require to run EXCAVATOR2.

NOTE: EXCAVATOR2 will run into problems if the target bed file contains duplicated or overlapping regions.

Preparation of analysis

Preparation of the analysis, and running EXCAVATOR2 can be done in any directory.

Make sure that perl, R, and samtools are loaded in your environment. Then create your EXCAVATOR2 analysis folder anywhere on your filesystem, e.g.:

mdkir EXCAVATOR2_Analysis_A
cd EXCAVATOR2_Analysis_A
mkdir preparation

You will then need to create a folder (EXCAVATOR2_Analysis_A/preparation/) for the output of the preparation step, and an input file for the preparation step.

The structure of the input file (Analysis_A_preparation_input.txt) should as follows:

path/to/sample1.bam path/to/EXCAVATOR2_Analysis_A/preparation/sample1 sample1
path/to/sample2.bam path/to/EXCAVATOR2_Analysis_A/preparation/sample2 sample2
path/to/sample3.bam path/to/EXCAVATOR2_Analysis_A/preparation/sample3 sample3
path/to/sample4.bam path/to/EXCAVATOR2_Analysis_A/preparation/sample4 sample4
path/to/sample5.bam path/to/EXCAVATOR2_Analysis_A/preparation/sample5 sample5
path/to/sample6.bam path/to/EXCAVATOR2_Analysis_A/preparation/sample6 sample6

In EXCAVATOR2_Analysis_A/ run:

perl /path/to/EXCAVATOR2_free/EXCAVATORDataPrepare.pl Analysis_A_preparation_input.txt \
 --processors 6 --target name-of-your-target_w50000 --assembly hg38

Running EXCAVATOR2

Once you have run the command to prepare the data for your analysis, you can call copy number changes using EXCAVATOR2 in your EXCAVATOR2_Analysis_A/ directory.
You will need to create an input file for EXCAVATOR2, the structure of which will differ depending on whether you will be running a paired or a pooled analysis.
For a pooled analysis where sample1 is a control sample, and samples 2-6 are all tumour samples, your input file EXCAVATOR2_input_pooled.txt would look as follows:

C1 path/to/EXCAVATOR2_Analysis_A/preparation/sample1 sample1
T1 path/to/EXCAVATOR2_Analysis_A/preparation/sample2 sample2
T2 path/to/EXCAVATOR2_Analysis_A/preparation/sample3 sample3
T3 path/to/EXCAVATOR2_Analysis_A/preparation/sample4 sample4
T4 path/to/EXCAVATOR2_Analysis_A/preparation/sample5 sample5
T5 path/to/EXCAVATOR2_Analysis_A/preparation/sample6 sample6

For a paired analysis where samples 1, 3, and 5 are a control samples that are matched to samples 2, 4, and 6, respectively, your input file EXCAVATOR2_input_paired.txt would look like this:

C1 path/to/EXCAVATOR2_Analysis_A/preparation/sample1 sample1
T1 path/to/EXCAVATOR2_Analysis_A/preparation/sample2 sample2
C2 path/to/EXCAVATOR2_Analysis_A/preparation/sample3 sample3
T2 path/to/EXCAVATOR2_Analysis_A/preparation/sample4 sample4
C3 path/to/EXCAVATOR2_Analysis_A/preparation/sample5 sample5
T3 path/to/EXCAVATOR2_Analysis_A/preparation/sample6 sample6

To run the paired analysis, make sure that perl, R, and samtools are loaded in your environment. Then run the following code inside the EXCAVATOR2_Analysis_A/ folder:

perl /path/to/EXCAVATOR2_free/EXCAVATORDataAnalysis.pl EXCAVATOR2_input_paired.txt \
 --processors 6 --target name-of-your-target_w50000 --assembly hg38 \
 --output /path/to/EXCAVATOR2_Analysis_A/results --mode paired

If you want to run the pooled analysis, run the following:

perl /path/to/EXCAVATOR2_free/EXCAVATORDataAnalysis.pl EXCAVATOR2_input_pooled.txt \
 --processors 6 --target name-of-your-target_w50000 --assembly hg38 \
 --output /path/to/EXCAVATOR2_Analysis_A/results --mode pooled

**Do not run more than one analysis in the same folder, e.g. do not run both a paired and a pooled analysis in EXCAVATOR2_Analysis_A/, create a separate folder for paired, and a separate folder for pooled. **

mjakobs / EXCAVATOR2_free