---
title: "Updated Proteomics Workflow"
output: html_notebook
---

Since others are going to start doing proteomics analysis, I figured I would take a bit of time to update the workflow, mostly to remove all of the installation and troubleshooting and to just show how to different programs work and interact.

As with other things I've done, this is all done in R, if you'd like to work through either command line, or Jupyter notebook, you can replace the system() calls with print(), which will cause R to just output the terminal commands to the console. 

Step 1: Convert .raw to .mzXML. 

```{r}

setwd("/home/shared/ReAdW")

system("WINEPREFIX=~/.wine32 wine ReAdW.2016010.msfilereader.exe ~/Documents/ProteomicsTest/20161205_sample_1.raw ~/Documents/ProteomicsTest/20161205_sample_1.mzXML")
```

Huzzah! We now have a file that works in Comet.

Step 2: Comet

I used a Gigas Proteome from Uniprot [found here](http://www.uniprot.org/proteomes/UP000005408) and a stock Comet params file from [here](http://comet-ms.sourceforge.net/parameters/parameters_201601/)

These files need to be in the same directory as your data files. The lack of spaces on the -P and -D arguments are also intentional.

```{r}
setwd("~/Documents/ProteomicsTest")

system("/home/shared/comet/comet.2016012.linux.exe -Pcomet.params.high-low -Dgigas-proteome.fasta 20161205_sample_1.mzXML")
```

Step 3: xinteract from the Trans-Proteomic Pipeline (TPP)

Arguments are from Emma's github post but Protein Prophet doesn't find any decoys with the tag "DECOY_", that may or may not be dataset/run specific?

```{r}

system("/usr/tpp_install/tpp/bin/xinteract -dDECOY_ -N20161205_sample_1 20161205_sample_1.pep.xml -p0.9 -OAp")
```

Step 4: Combine the different interact files via Protein Prophet

This comes from the UWPR Abacus page found [here](http://proteomicsresource.washington.edu/protocols06/Abacus/).

```{r}
 
system("/usr/tpp_install/tpp/bin/ProteinProphet interact*.pep.xml interact-COMBINED.prot.xml")

```

Step 5: Abacus

This part is a little sketchy, as it requires a parameters file that is unique to each run (It sets up the directory structure used). I posted a sample one I used for this particular test and one can change it to fit their needs, or one can run the .jar file with the GUI (requires physical access to the machine/a VNC connection) to create your own parameters file. [File here](https://raw.githubusercontent.com/seanb80/seanb80.github.io/master/files/Abacus_parameters.txt)

```{r}

system("java -Xmx16g -jar /home/shared/abacus/abacus.jar -p ~/Documents/ProteomicsTest/Abacus_parameters.txt")
```

```{r}
system("head ABACUS_output.tsv")
```

Well, It worked. I think?