[](https://travis-ci.org/pliu55/pram)
[](http://bioconductor.org/packages/devel/bioc/html/pram.html)
PRAM: Pooling RNA-seq and Assembling Models
===========================================
Table of Contents
-----------------
* [Introduction](#Introduction)
* [Installation](#Installation)
* [Reference](#Reference)
* [Contact](#Contact)
* [License](#License)
* * *
## <a name='Introduction'></a> Introduction
Pooling RNA-seq and Assembling Models (__PRAM__) is an __Bioconductor__ __R__
package that
utilizes multiple RNA-seq datasets to predict transcript models. The workflow
of PRAM contains four steps, which is shown in
the figure below with function names and associated key parameters. PRAM has a
[vignette](https://bioconductor.org/packages/devel/bioc/vignettes/pram/inst/doc/pram.pdf) that describes each function in details.
<p align='center'>
<img src="vignettes/workflow_noScreen.jpg" width="400" height="407">
</p>
## <a name='Installation'></a> Installation
### From GitHub
Start __R__ and enter:
```r
devtools::install_github('pliu55/pram')
```
### From Bioconductor
Start __R__ and enter:
```r
if (!requireNamespace("BiocManager", quietly = TRUE))
install.packages("BiocManager")
BiocManager::install("pram")
```
<!--
- Cufflinks v2.2.1 macOS binary have some issues
- it will report segmentation fault for the same bam file, which Linux
Cufflinks runs ok
- Have to use Cufflinks v2.1.1 for macOS instead
-->
<!--
## <a name='Quick-start'></a>Quick start
PRAM provides a function `runPRAM()` to let you run through the whole workflow.
### <a name='predict-only'></a> Predict transcript models only
For a given gene annotation and RNA-seq alignments, you can predict transcript
models in intergenic genomic regions:
```R
runPRAM(in_gtf, in_bamv, out_gtf)
```
- `in_gtf`: an input GTF file defining genomic coordinates of existing genes.
Required to have an attribute of __gene_id__ in the ninth column.
- `in_bamv`: a vector of input BAM file(s) containing RNA-seq alignments.
Currently,
PRAM only supports strand-specific paired-end RNA-seq with the
first mate on the right-most of transcript coordinate, i.e.,
'fr-firststrand' by Cufflinks definition.
- `out_gtf`: an output GTF file of predicted transcript models
### <a name='predict-screen'></a> Predict transcript models and screen them by ChIP-seq
If you are interested to predict intergenic transcripts for a particular cell
or tissue type, you can use epigenetic ChIP-seq
data together with known transcripts and their expression levels to further
screen intergenic transcript models:
```
runPRAM(in_gtf, in_bamv, out_gtf, in_bedv, training_tpms, training_gtf)
```
- `in_gtf`, `in_bamv`, and `out_gtf` are the same as described above
- `in_bedv`: A vector of BED file(s) containing ChIP-seq alignments.
- `training_tpms`: A vector of RSEM quantification results for known
transcripts
- `training_gtf`: A GTF file defining genomic coordinates of known
transcripts
### <a name='Examples'></a> Examples
PRAM has included input examples files in its `extdata/demo/`
folder. The table below provides a quick summary of all the example files.
| input argument | file name(s) |
|:--------------:|:------------:|
| `in_gtf` | [in.gtf](inst/extdata/demo/in.gtf) |
| `in_bamv` | [SZP.bam](inst/extdata/demo/SZP.bam), [TLC.bam](inst/extdata/demo/TLC.bam) |
| `in_bedv` | H3K79me2.bed.gz, POLR2.bed.gz |
| `training_tpms`| AED1.isoforms.results, AED2.isoforms.results |
| `training_gtf` | training.gtf |
You can access example files by `system.file()` in __R__, e.g. for the
argument `in_gtf`, you can access its example file by
```R
system.file('extdata/demo/in.gtf', package='pram')
```
Below shows usage of `runPRAM()` with example input files:
##
## Predict transcript models only
##
```R
in_gtf = system.file('extdata/demo/in.gtf', package='pram')
in_bamv = c( system.file('extdata/demo/SZP.bam', package='pram'),
system.file('extdata/demo/TLC.bam', package='pram') )
pred_out_gtf = tempfile(fileext='.gtf')
runPRAM(in_gtf, in_bamv, pred_out_gtf)
```
##
## Predict transcript models and screen them by ChIP-seq data
##
in_bedv = c( system.file('extdata/demo/H3K79me2.bed.gz', package='pram'),
system.file('extdata/demo/POLR2.bed.gz', package='pram') )
training_tpms = c( system.file('extdata/demo/AED1.isoforms.results', package='pram'),
system.file('extdata/demo/AED2.isoforms.results', package='pram') )
training_gtf = system.file('extdata/demo/training.gtf', package='pram')
screen_out_gtf = tempfile(fileext='.gtf')
runPRAM(in_gtf, in_bamv, screen_out_gtf, in_bedv, training_tpms, training_gtf)
-->
## <a name="Reference"></a> Reference
__PRAM: a novel pooling approach for discovering intergenic transcripts from large-scale RNA sequencing experiments__. Peng Liu, Alexandra A. Soukup, Emery H. Bresnick, Colin N. Dewey, and Sündüz Keleş. _bioRxiv_, 2019. __doi__: https://doi.org/10.1101/636282
For key results reported in the PRAM manuscript and scripts for
reproducibility, please check out
[this GitHub repository](https://github.com/pliu55/pram_paper).
## <a name="Contact"></a> Contact
Got a question? Please report it at the [issues tab](https://github.com/pliu55/pram/issues) in this repository.
## <a name="License"></a> License
PRAM is licensed under the [GNU General Public License v3](LICENSE).
