---
title: 'AIRR Data Representation Reference Library Usage'
author: "The AIRR Community"
date: '`r Sys.Date()`'
output:
  html_document:
    standalone: true
    fig_height: 4
    fig_width: 7.5
    highlight: pygments
    theme: readable
    toc: yes 
  md_document:
    fig_height: 4
    fig_width: 7.5
    preserve_yaml: no
    toc: yes
  pdf_document:
    dev: pdf
    fig_height: 4
    fig_width: 7.5
    highlight: pygments
    toc: yes
geometry: margin=1in
fontsize: 11pt
vignette: >
  %\VignetteIndexEntry{Usage Vignette}
  %\usepackage[utf8]{inputenc}
  %\VignetteEngine{knitr::rmarkdown}
---

## Introduction

Since the use of High-throughput sequencing (HTS) was first introduced to analyze 
immunoglobulin (B-cell receptor, antibody) and T-cell receptor repertoires 
(Freeman et al, 2009; Robins et al, 2009; Weinstein et al, 2009), 
the increasing number of studies making use of this technique has produced enormous 
amounts of data and there exists a pressing need to develop and adopt common standards, 
protocols, and policies for generating and sharing data sets. The [Adaptive Immune 
Receptor Repertoire (AIRR) Community](http://airr-community.org) formed in 
2015 to address this challenge (Breden et al, 2017) and has stablished the set of 
minimal metadata elements (MiAIRR) required for describing published AIRR 
datasets (Rubelt et al, 2017) as well as file formats to represent this data in a 
machine-readable form. The `airr` R package provide read, write and validation of data 
following the AIRR Data Representation schemas. This vignette provides a set of 
simple use examples.

### AIRR Data Standards

The AIRR Community's recommendations for a minimal set of metadata that should be 
used to describe an AIRR-seq data set when published or deposited in a 
AIRR-compliant public repository are described in Rubelt et al, 2017. The primary 
aim of this effort is to make published AIRR datasets FAIR (findable, accessible, 
interoperable, reusable); with sufficient detail such that a person skilled in the 
art of AIRR sequencing and data analysis will be able to reproduce the 
experiment and data analyses that were performed.

Following this principles, V(D)J reference alignment annotations are 
saved in standard tab-delimited files (TSV) with associated metadata 
provided in accompanying YAML formatted files. The column names 
and field names in these files have been defined by the 
AIRR Data Representation Working Group using a controlled vocabulary of 
standardized terms and types to refer to each piece of information.

## Reading AIRR formatted files

The `airr` package contains the function `read_rearrangement` to read and validate
files containing AIRR Rearrangement records, where a Rearrangement record describes 
the collection of optimal annotations on a single sequence that has undergone V(D)J reference alignment. 
The usage is straightforward, as the file format is a typical tabulated file. 
The argument that needs attention is `base`, with possible values `"0"` and `"1"`. 
`base` denotes the starting index for positional fields in the input file. 
Positional fields are those that contain alignment coordinates and names 
ending in "_start" and "_end". If the input file is using 1-based closed 
intervals (R style), as defined by the standard, then positional fields will 
not be modified under the default setting of `base="1"`. If the input file is using 
0-based coordinates with half-open intervals (python style), then 
positional fields may be converted to 1-based closed 
intervals using the argument `base="0"`.

### Reading Rearrangements
```{r}
# Imports
library(airr)
library(tibble)

# Read Rearrangement example file
f1 <- system.file("extdata", "rearrangement-example.tsv.gz", package="airr")
rearrangement <- read_rearrangement(f1)
glimpse(rearrangement)
```

### Reading AIRR Data Models

AIRR Data Model records, such as Repertoire and GermlineSet, can be read from either a YAML 
or JSON formatted file into a nested list.

```{r}
# Read Repertoire example file
f2 <- system.file("extdata", "repertoire-example.yaml", package="airr")
repertoire <- read_airr(f2)
glimpse(repertoire)

# Read GermlineSet example file
f3 <- system.file("extdata", "germline-example.json", package="airr")
germline <- read_airr(f3)
glimpse(germline)
```


## Writing AIRR formatted files

The `airr` package contains the function `write_rearrangement` to write
Rearrangement records to the AIRR TSV format.

### Writing Rearrangements

```{r}
x1 <- file.path(tempdir(), "airr_out.tsv")
write_rearrangement(rearrangement, x1)
```

### Writing AIRR Data Models

AIRR Data Model records can be written to either YAML or JSON using the `write_airr` function.

```{r}
x2 <- file.path(tempdir(), "airr_repertoire_out.yaml")
write_airr(repertoire, x2, format="yaml")

x3 <- file.path(tempdir(), "airr_germline_out.json")
write_airr(germline, x3, format="json")
```

## Validating AIRR data structures

The `airr` package contains the function `validate_rearrangement` to validate
tabular (`data.frame`) Rearrangement records and AIRR Data Model objects, respectively.

```{r}
# Validate Rearrangement data.frame
validate_rearrangement(rearrangement)

# Validate an AIRR Data Model
validate_airr(repertoire)

# Validate AIRR Data Model records individual 
validate_airr(germline, each=TRUE)
```

## References

1. Breden, F., E. T. Luning Prak, B. Peters, F. Rubelt, C. A. Schramm, C. E. Busse, 
    J. A. Vander Heiden, et al. 2017. Reproducibility and Reuse of Adaptive Immune 
    Receptor Repertoire Data. *Front Immunol* 8: 1418.
2. Freeman, J. D., R. L. Warren, J. R. Webb, B. H. Nelson, and R. A. Holt. 2009. 
    Profiling the T-cell receptor beta-chain repertoire by massively parallel sequencing.
    *Genome Res* 19 (10): 1817-24.
3. Robins, H. S., P. V. Campregher, S. K. Srivastava, A. Wacher, C. J. Turtle, 
    O. Kahsai, S. R. Riddell, E. H. Warren, and C. S. Carlson. 2009. 
    Comprehensive assessment of T-cell receptor beta-chain diversity in alphabeta T cells.
    *Blood* 114 (19): 4099-4107.
4. Rubelt, F., C. E. Busse, S. A. C. Bukhari, J. P. Burckert, E. Mariotti-Ferrandiz, 
    L. G. Cowell, C. T. Watson, et al. 2017. Adaptive Immune Receptor Repertoire Community 
    recommendations for sharing immune-repertoire sequencing data.
    *Nat Immunol* 18 (12): 1274-8.
5. Weinstein, J. A., N. Jiang, R. A. White, D. S. Fisher, and S. R. Quake. 2009.
    High-throughput sequencing of the zebrafish antibody repertoire.
    *Science* 324 (5928): 807-10.