Avant-garde

Avant-garde (AvG) is a new tool to refine Data-Independent Acquisition (and Parallel Reaction Monitoring) by removing interfered transitions, adjusting integration boundaries and scoring peaks to control the FDR. Unlike other tools where MS runs are scored independently from each other, Avant-garde uses a novel data-driven scoring strategy. DIA signals are refined by learning from the data itself, using all measurements in all samples together to achieve the best optimization. Avant-garde evaluates the suitability of a peak to be used for quantification. It is capable of improving the selectivity, accuracy, and reproducibility of the quantification results in very complex biological matrices, reachng the same levels obtained with manual validation. AvG is described in detail in our bioRxiv paper.

Getting Started

• Avant-garde is an R-based package that works with Skyline's reports.
• This github repository contains a tutorial decribing how to install, how to run the package as an External tool in Skyline and how to import the results back into Skyline.

Installation of the Skyline external tool

The installation can be done directly from Skyline.

1. In Skyline go to Tools.
2. Select Tool Store.
3. Select AvantGardeDIA.
4. Click Install.

Installation from the Skyline Tool Store.

Download the zip file and follow instructions here.

Installation from this repo

1. Download the latest zip file here.
2. Follow instructions found in the tutorial.

The installation should take 1 to 2 minutes depending on the system configuration. The Skyline External tool will automatically install the Avant-garde R package

Installation of the R package

From tar.gz file

1. Download latest tar.gz file located here.
2. Run this code and point to the location of the tar.gz file.

install.packages(pkgs = file.choose(), repos = NULL, type = "source")

Using devtools

1. install the devtools package.

install.packages("devtools")

2. Load the devtools package.

library(devtools)

3. Install the package using 'install_github'.

install_github("SebVaca/Avant_garde_Publication", subdir="r_package/AvantGardeDIA")

Built With

• Skyline - Windows client application for building quantitative proteomic methods and analyzing mass spectrometer data.
• GA - R package for genetic algorithms.
• tidyverse - Tidyverse collection of R packages for data science.

Software versions

The Avant-garde package functions with the following software versions:

Skyline (>= 4.2.0)
R (>=3.5.2)

R Package Versions

The Avant-garde package functions with the following R package versions:

GA (>= 3.0.2)
doSNOW (>= 1.0.16)
snow (>= 0.4-2)
iterators (>= 1.0.9)
foreach (>= 1.4.4)
sqldf (>= 0.4-11)
RSQLite (>= 2.0)
gsubfn (>= 0.6-6)
proto (>= 1.0.0)
data.table (>= 1.10.4-3)
stringr (>= 1.3.0)
tidyr (>= 0.8.0)
dplyr (>= 0.7.4)

Support

If you have a question about using AvG, or if you encounter a problem, you can post your questions here.

Demo

A demo data set can be found here.

• The demo contains a Skyline file, instructions to run AvG on the demo and a folder containing the expected results.
• The expected results are in the Demo_results.zip file. It contains all the output reports and the Skyline file after the data curation by AvG.
• The demo requires Skyline (version >= 4.2.0) and R (version >=3.5.2).
• The demo should take 6 minutes to run on a system with similar configuration as the following: Windows 7, Intel Core i7-3770 CPU @ 3.40Ghz 3.40 GHz, memory (RAM) 24Gb, 64-bit operating system.

Authors

• Sebastian Vaca - Broad Institute, Cambridge, MA
• Jacob D. Jaffe - Broad Institute, Cambridge, MA

Acknowledgments

• Skyline development team
• Nick Schulman, Brendan MacLean and Michael MacCoss - for their help integrating Avant-garde into Skyline.