MSnbase, efficient and elegant R-based processing of raw mass spectrometry data

class: center, middle, inverse, title-slide

# MSnbase, efficient and elegant R-based processing of raw mass spectrometry data
### Laurent Gatto (UCLouvain, Belgium), Sebastian Gibb (University of Greifswald, Germany) and Johannes Rainer (Eurac Research, Italy)
### 15 July 2020

---

class: inverse, middle

## What is MSnbase

- **R/[Bioconductor](http://bioconductor.org/packages/release/bioc/html/MSnbase.html)
  package**, publicly available since **October 2010**.
- Developed and maintained by a **core team** of developers (Laurent
  Gatto, Sebastian Gibb, Johannes Rainer) + 9 additional [official
  contributors](https://lgatto.github.io/msnbase-contribs-2/).
- Thoroughly **tested** and **documented**: see http://lgatto.github.io/MSnbase/

#### **TODAY**: focus on raw mass spectrometry data

Details in

> *MSnbase, efficient and elegant R-based processing and visualisation
> of raw mass spectrometry data* Laurent Gatto, Sebastian Gibb,
> Johannes Rainer bioRxiv 2020.04.29.067868; doi:
> https://doi.org/10.1101/2020.04.29.067868

---
class: middle, center, inverse

## What do we mean by *raw data*?

---
class: center, middle
background-image: url("./figs/raw.png")
background-size: contain

---
class: middle

```r
> library("MSnbase") ## load the package
> f <- msdata::proteomics(pattern = "2014", full.names = TRUE)
> basename(f) ## example file 
[1] "TMT_Erwinia_1uLSike_Top10HCD_isol2_45stepped_60min_01-20141210.mzML.gz"
*> in_mem <- readMSData(f, mode = "inMemory") ## in-memory backend
*> on_disk <- readMSData(f, mode = "onDisk") ## on-disk backend
> on_disk ## R representation of an MS experiment
MSn experiment data ("OnDiskMSnExp")
Object size in memory: 3.31 Mb
- - - Spectra data - - -
 MS level(s): 1 2
 Number of spectra: 7534
 MSn retention times: 0:0 - 60:2 minutes
- - - Processing information - - -
Data loaded [Tue Jun 30 16:24:33 2020]
 MSnbase version: 2.14.2
[...]
```

### -> Comparison of the **in-memory** and **on-disk** _backends_

---
class: middle, center, inverse

## Results

---
class: center, middle
background-image: url("./figs/res.png")
background-size: contain

---
class: middle
```r
> on_disk <- on_disk %>% ## keep only
 filterMsLevel(2) %>% ## MS2 spectra
 filterRt(c(1000, 3000)) %>% ## acquired 1000 and 3000 seconds
 filterMz(120, 135) ## m/z between 120 and 135
## keep 50% most intense spectra
> on_disk[precursorIntensity(on_disk) > 11e6]
```

---
class: center, middle
background-image: url("./figs/res.png")
background-size: contain

---
class: middle
name: laurent-gatto

.left-col-50[
<img src="./img/lgatto3b.png" width = "180px"/>
### Laurent Gatto
&nbsp;[Computational Biology Group](https://lgatto.github.io/cbio-lab/) 
&nbsp;de Duve Institute, UCLouvain 
&nbsp;laurent.gatto@uclouvain.be 
&nbsp;https://lgatto.github.io 
&nbsp;[@lgatto](https://twitter.com/lgatt0/) 
&nbsp;[lgatto](https://github.com/lgatto/) 
]

.rigth-col-50[

## Want to learn more

**Slides** https://lgatto.github.io/2020_07_15_ISMB/

**Pre-print** https://www.biorxiv.org/content/10.1101/2020.04.29.067868v3

`MSnbase` **webpage** http://lgatto.github.io/MSnbase/

## Thank you for your attention

]