Plot replicate correlation

usage: replicate_correlation.py [-h] [-j JID] -f INPUT_LIST [--bw]
                                [--featureCount_addon_parameters FEATURECOUNT_ADDON_PARAMETERS]

optional arguments:
  -h, --help            show this help message and exit
  -j JID, --jid JID     enter a job ID, which is used to make a new directory.
                        Every output will be moved into this folder. (default:
                        replicate_correlation_yli11_2020-04-20)
  -f INPUT_LIST, --input_list INPUT_LIST
                        tsv 4 columns, bam 1, bam 2, peak 1, peak 2. Relative
                        or Absolute path. (default: None)
  --bw                  input is bw, not bam (default: False)
  --featureCount_addon_parameters FEATURECOUNT_ADDON_PARAMETERS
                        if paired data add -p option (default: )

Summary

This is a scatter plot showing scatters, density, and correlation R square.

bed file should have 4 columns: chr, start, end, name

Input

As of 4-20-2020, the input format now also accept bw files.

Option 1:

A tsv file with 4 columns (2 bam files and 2 peak files): bam 1, bam 2, peak 1, peak 2.

Option 2:

A tsv file with 4 columns (2 bw files and 2 peak files): bw 1, bw 2, peak 1, peak 2.

Usage

hpcf_interactive

module load python/2.7.13

replicate_correlation.py -f input.list

For paired-end data use (bam input):

replicate_correlation.py -f input.list --featureCount_addon_parameters " -p"

For bw input (no difference SE or PE)

replicate_correlation.py -f input.list --bw

correlation on intersect sets

replicate_correlation.py -f correlation.list --bw --intersect -j intersection
replicate_correlation.py -f correlation.list --bw  -j union

Output

The output file will be emailed to you.

Highlight some dots

The script we are using is from : scatter_correlation.py. Specifically, the usage for the output from replicate_correlation.py is as follows.

1. Input: signal.out

Find the signal.out file in the output jid folder. This is a tsv file, open it in excel, edit the column names to what you need. The first column is the index, by default the column name is “0”. You don’t need to modify this name. The last two columns are used as X-axis and Y-axis name, change them to what you need, spaces are allowed.

Next, find the row where you want to highlight, modify its index name to something simple, you will need to input this name later.

Example input:

0       WT      KO
test    150.76201       4.45338
Banana_2,Banana_2       7.28372 7.71828
Banana_3,Banana_3       6.638380000000001       5.2838400000000005
Banana_4,Banana_4       6.5786  7.52588
Banana_5,Banana_5       7.51205 6.96104
Banana_6,Banana_6       5.625030000000001       3.6350599999999997
Banana_7,Banana_7       5.18235 3.84519
Banana_8,Banana_8       5.73767 5.34589
Banana_9,Banana_9       7.53337 3.71176

2. Run scatter_correlation.py

hpcf_interactive

module load conda3

source activate /home/yli11/.conda/envs/py2

scatter_density.py -f signal.out -s "\t" -x WT -y KO --index 0 --highlight test -o myoutput.pdf

3. example figure

code @ github.