Skip to content

Examples.md

Latest commit

 

History

History
63 lines (43 loc) · 3.31 KB

Examples.md

File metadata and controls

63 lines (43 loc) · 3.31 KB

Examples

All sample datasets can be downloaded from the Downloads page.

Sample input

PRAS requires a binding table, an annotation table, and an ID table as input files (described in the Instructions). Here, we present a set of toy datasets as an example for PRAS usage. The binding table is a subset of the reliable cross-linking sites from CELF4 iCLIP-seq dataset used in our paper. The annotation table is the GTF file of Refseq mm9 annotation. The ID file is the transcript IDs and gene names of Refseq mm9 annotation. The files looks like the screenshots as below:

1. Sample binding table: test_binding_tab.bed

alt text

Note that the fifth column could have sign to indicate the strand.

2. Sample annotation table: mm9_refseq.gtf

alt text

3. Sample ID table: mm9_refseq_ID.txt

alt text

In addition, PRAS support calculating the scores on a customized gene list via -l option. If it's not given, PRAS will perform the calculation on all the genes in the annotation files.

Command example

1. run PRAS "check" mode to examine the binding profiles around the candidate reference sites.

python PRAS_1.0.py -g mm9_refseq.gtf -t mm9_refseq_id.txt -i test_binding_tab.bed -m check -s transcript -a test_binding_tab_assign.txt -w 10 -c 500 -p check_test

2. run PRAS with known reference site.

python PRAS_1.0.py -g mm9_refseq.gtf -t mm9_refseq_id.txt -m score -s 3UTR -i test_binding_tab.bed -a test_binding_tab_assign.txt -w 10 -r 3 -d 1000

Sample output

1. Sample output table 1: test_binding_tab_utr3.assign.txt

In "check" mode, peaks in 5' UTR, CDS and 3' UTR are all annotated. In "score" mode, only peaks in selected genomic region are annotated. The format of the annotated peak file is consistent for the two modes, which is shown as below.

alt text

2. Sample output plot: check_test_500nt_around_TSS_TIS_TTS_TES.pdf

If PRAS "check" mode is enabled, the user will get the binding profile plot around the candidate reference sites. Here is the output plot of the toy dataset when -s transcript is enabled.

alt text

We can see the 3'UTR enrichment from the binding profile plots. In addition, PRAS also allows splice site binding check via -s splice option. Users can try it using RBPs with splicing-related function.

3. Sample output suggestion file: check_test_500nt_PRAS_option_suggestions.txt

If PRAS "check" mode is enabled, the user will get the parameter suggestion file based on the binding profile plot around the candidate reference sites. These parameters can be used in the "score" mode if there is no pre-knowledge of the target RBP.

alt text

4. Sample output table: binding_PRAS.txt

Given known reference site, PRAS will generate a score table with four columns which are gene name, PRAS score, log2 of PRAS score, and the rank of the gene.

alt text

One will get something similar to the following lines if PRAS run successfully.

If PRAS "check" option is enabled, the process is as below:

alt text

If PRAS is run with known reference site in "score" mode, the process is as below:

alt text