---
title: "Description and usage of MsBackendMgf"
output:
    BiocStyle::html_document:
        toc_float: true
vignette: >
    %\VignetteIndexEntry{Description and usage of MsBackendMgf}
    %\VignetteEngine{knitr::rmarkdown}
    %\VignetteEncoding{UTF-8}
    %\VignettePackage{MsBackendMgf}
    %\VignetteDepends{Spectra,BiocStyle,BiocParallel}
---

```{r style, echo = FALSE, results = 'asis', message=FALSE}
BiocStyle::markdown()
```

**Package**: `r Biocpkg("MsBackendMgf")`<br />
**Authors**: `r packageDescription("MsBackendMgf")[["Author"]] `<br />
**Last modified:** `r file.info("MsBackendMgf.Rmd")$mtime`<br />
**Compiled**: `r date()`

```{r, echo = FALSE, message = FALSE}
library(Spectra)
library(BiocStyle)
```

# Introduction

The `r Biocpkg("Spectra")` package provides a central infrastructure for the
handling of Mass Spectrometry (MS) data. The package supports interchangeable
use of different *backends* to import MS data from a variety of sources (such as
mzML files). The `r Biocpkg("MsBackendMgf")` package allows the import of MS/MS
data from MGF ([Mascot Generic
Format](http://www.matrixscience.com/help/data_file_help.html)) files. The
`MsBackendMgf` backend allows to load and represent data from these *standard*
MGF files, while the `MsBackendAnnotatedMgf` backend supports import of data
from MGF files containing also annotations for the individual mass peaks. This
vignette illustrates the usage of the *MsBackendMgf* package.

# Installation

To install this package, start `R` and enter:

```{r, eval = FALSE}
if (!requireNamespace("BiocManager", quietly = TRUE))
    install.packages("BiocManager")

BiocManager::install("MsBackendMgf")
```

This will install this package and all eventually missing dependencies.


# Importing MS/MS data from MGF files

MGF files store one to multiple spectra, typically centroided and of MS
level 2. In our short example below, we load MGF files which are provided with
this package. In a next section we also import data from an MGF file containing
individual peak annotations. Below we first load all required packages and
define the paths to the MGF files.

```{r load-libs}
library(Spectra)
library(MsBackendMgf)

fls <- dir(system.file("extdata", package = "MsBackendMgf"),
           full.names = TRUE, pattern = "^spectra(.*).mgf$")
fls
```

MS data can be accessed and analyzed through `Spectra` objects. Below
we create a `Spectra` with the data from these MGF files. To this end
we provide the file names and specify to use a `MsBackendMgf()`
backend as *source* to enable data import.

```{r import}
sps <- Spectra(fls, source = MsBackendMgf())
```

With that we have now full access to all imported spectra variables
that we list below.

```{r spectravars}
spectraVariables(sps)
```

Besides default spectra variables, such as `msLevel`, `rtime`,
`precursorMz`, we also have additional spectra variables such as the
`TITLE` of each spectrum in the MGF file.

```{r instrument}
sps$rtime
sps$TITLE
```

By default, fields in the MGF file are mapped to spectra variable names using
the mapping returned by the `spectraVariableMapping()` function:

```{r spectravariables}
spectraVariableMapping(MsBackendMgf())
```

The names of this `character` vector are the spectra variable names (such as
`"rtime"`) and the field in the MGF file that contains that information are the
values (such as `"RTINSECONDS"`). Note that it is also possible to overwrite
this mapping (e.g. for certain MGF *dialects*) or to add additional
mappings. Below we add the mapping of the MGF field `"TITLE"` to a spectra
variable called `"spectrumName"`.

```{r map}
map <- c(spectrumName = "TITLE", spectraVariableMapping(MsBackendMgf()))
map
```

We can then pass this mapping to the `backendInitialize()` method, or the
`Spectra()` constructor.

```{r import2}
sps <- Spectra(fls, source = MsBackendMgf(), mapping = map)
```

We can now access the spectrum's title with the newly created spectra variable
`"spectrumName"`:

```{r spectrumName}
sps$spectrumName
```

In addition we can also access the m/z and intensity values of each
spectrum.

```{r mz}
mz(sps)
intensity(sps)
```

The `MsBackendMgf` backend allows also to export data in MGF format. Below we
export the data to a temporary file. We hence call the `export()` function on
our `Spectra` object specifying `backend = MsBackendMgf()` to use this backend
for the export of the data. Note that we use again our custom mapping of
variables such that the spectra variable `"spectrumName"` will be exported as
the spectrums' title.

```{r export}
fl <- tempfile()
export(sps, backend = MsBackendMgf(), file = fl, mapping = map)
```

We next read the first lines from the exported file to verify that the title was
exported properly.

```{r export-check}
readLines(fl)[1:12]
```

Note that the `MsBackendMgf` exports all spectra variables as fields in the mgf
file. To illustrate this we add below a new spectra variable to the object and
export the data.

```{r}
sps$new_variable <- "A"
export(sps, backend = MsBackendMgf(), file = fl)
readLines(fl)[1:12]
```

We can see that also our newly defined variable was exported. Also, because we
did not provide our custom variable mapping this time, the variable
`"spectrumName"` was **not** used as the spectrum's title.

Sometimes it might be required to not export all spectra variables since some
exported fields might not be recognized/supported by external tools. Using the
`selectSpectraVariables()` function we can reduce our `Spectra` object to export
to contain only relevant spectra variables. Below we restrict the data to only
m/z, intensity, retention time, acquisition number, precursor m/z and precursor
charge and export these to an MGF file. Also, some external tools don't support
the `"TITLE"` field in the MGF file. To disable export of the spectrum ID/title
`exportTitle = FALSE` can be used.

```{r}
sps_ex <- selectSpectraVariables(sps, c("mz", "intensity", "rtime",
                                        "acquisitionNum", "precursorMz",
                                        "precursorCharge"))
export(sps_ex, backend = MsBackendMgf(), file = fl, exportTitle = FALSE)
readLines(fl)[1:12]
```

## Annotated MGF files

Variants of MGF files can also contain annotations for the individual mass
peaks. These files are expected to contain one line per mass peak, with the
first two elements being the mass peak's *m/z* and intensity values. All
(eventually present) additional elements are considered *annotations*. The
`MsBackendAnnotatedMgf` backend can be used to import and represent such files.
Below we read the first 10 lines of an example annotated MGF file.

```{r}
fl <- dir(system.file("extdata", package = "MsBackendMgf"),
          pattern = "fiora", full.names = TRUE)
readLines(fl, 10)
```

The last 3 displayed lines show the mass peak content of a MGF spectrum with the
*m/z*, intensity and an annotation for the mass peak. Such data files can be
imported using the `MsBackendAnntotatedMgf` backend:

```{r}
sps_ann <- Spectra(fl, source = MsBackendAnnotatedMgf())
sps_ann
```

The peaks annotation have also been imported and are available as additional
*peaks variables*:

```{r}
peaksVariables(sps_ann)
```

The names of the annotations follow the standard R convention of naming columns
in a `data.frame` if no name is provided, i.e. it consists of `"V"` followed by
a number. The data can be retrieved with the `peaksData()` function specifying
the names of the peaks variables to extract. If not provided, `peakData()` will
only extract the *m/z* and intensity values. Thus, below we call `peaksData()`
providing the names of all requested variables (columns).

```{r}
pd <- peaksData(sps_ann, columns = c("mz", "intensity", "V1"))
```

The peaks data of one spectrum is now represented as a `data.frame`:

```{r}
pd[[1L]] |>
    head()
```

# Parallel processing

The *MsBackendMgf* package supports parallel processing for data
import. Parallel processing can be enabled by providing the parallel processing
setup to the `backendInitialize()` function (or the `readMgf()` function) with
the `BPPARAM` parameter. By default (with `BPPARAM = SerialParam()`) parallel
processing is disabled. If enabled, and data import is performed on a single
file, the extraction of spectra information on the imported MGF file is
performed in parallel. If data is to be imported from multiple files, the import
is performed in parallel on a per-file basis (i.e. in parallel from the
different files). Generally, the performance gain through parallel processing is
only moderate and it is only suggested if a large number of files need to be
processed, or if the MGF file is very large (e.g. containing over 100,000
spectra).

# Session information

```{r}
sessionInfo()
```