README.md

In VitorAguiar/hlahaps: Retrieve HLA haplotype information from NMDP data.

README

Install

First, we need to install devtools, if not already installed.

> install.packages("devtools")

Then, we install hlahaps using the install_github() function.

> devtools::install_github("VitorAguiar/hlahaps")

Now, we can load hlahaps as any other R package using the base function library().

> library(hlahaps)

Usage

When we load the package, the Gourraud et al. (2014) data is loaded.

> pag

# A tibble: 1,910 × 5
   subject        A        C        B       DRB1
     <chr>    <chr>    <chr>    <chr>      <chr>
1  HG00096 A*01:01g C*07:01g B*08:01g DRB1*03:01
2  HG00096  A*29:02  C*16:01  B*44:03 DRB1*07:01
3  HG00097 A*03:01g C*07:02g B*07:02g DRB1*13:03
4  HG00097 A*24:02g C*07:02g B*07:02g DRB1*15:01
5  HG00099 A*68:01g C*07:04g B*44:02g DRB1*03:01
6  HG00099 A*01:01g C*07:01g B*08:01g DRB1*11:01
7  HG00100 A*01:01g C*07:01g B*08:01g DRB1*03:01
8  HG00100 A*01:01g  C*06:02  B*57:01 DRB1*07:01
9  HG00101  A*32:01  C*02:02 B*27:05g DRB1*04:04
10 HG00101 A*11:01g  C*06:02  B*57:01 DRB1*15:01
# ... with 1,900 more rows

We will use the individual HG00096 to demonstrate how the query_nmdp() function works:

> test_ind <- dplyr::filter(pag, subject == "HG00096")
> test_ind

# A tibble: 2 × 5
  subject        A        C        B       DRB1
    <chr>    <chr>    <chr>    <chr>      <chr>
1 HG00096 A*01:01g C*07:01g B*08:01g DRB1*03:01
2 HG00096  A*29:02  C*16:01  B*44:03 DRB1*07:01

Applying query_nmdp():

> query_nmdp(test_ind)

$`original haplotypes`
# A tibble: 2 × 4
         A        C        B       DRB1
     <chr>    <chr>    <chr>      <chr>
1 A*01:01g C*07:01g B*08:01g DRB1*03:01
2  A*29:02  C*16:01  B*44:03 DRB1*07:01

$`haplotypes found`
# A tibble: 2 × 14
         A        C        B       DRB1    AFA_freq AFA_rank     API_freq
     <chr>    <chr>    <chr>      <chr>       <dbl>    <int>        <dbl>
1 A*01:01g C*07:01g B*08:01g DRB1*03:01 0.010938011        2 0.0011127143
2 A*29:02g  C*16:01  B*44:03 DRB1*07:01 0.003515995       14 0.0002314976
# ... with 7 more variables: API_rank <int>, CAU_freq <dbl>,
#   CAU_rank <int>, HIS_freq <dbl>, HIS_rank <int>, NAM_freq <dbl>,
#   NAM_rank <int>

$`haplotypes not found`
# A tibble: 0 × 4
# ... with 4 variables: A <chr>, C <chr>, B <chr>, DRB1 <chr>

$`possible haplotypes`
# A tibble: 14 × 4
          A        C        B       DRB1
      <chr>    <chr>    <chr>      <chr>
1  A*01:01g C*07:01g B*08:01g DRB1*03:01
2  A*01:01g C*07:01g B*08:01g DRB1*07:01
3  A*01:01g C*07:01g  B*44:03 DRB1*03:01
4  A*01:01g C*07:01g  B*44:03 DRB1*07:01
5  A*01:01g  C*16:01 B*08:01g DRB1*03:01
6  A*01:01g  C*16:01  B*44:03 DRB1*03:01
7  A*01:01g  C*16:01  B*44:03 DRB1*07:01
8  A*29:02g C*07:01g B*08:01g DRB1*03:01
9  A*29:02g C*07:01g B*08:01g DRB1*07:01
10 A*29:02g C*07:01g  B*44:03 DRB1*03:01
11 A*29:02g C*07:01g  B*44:03 DRB1*07:01
12 A*29:02g  C*16:01 B*08:01g DRB1*03:01
13 A*29:02g  C*16:01  B*44:03 DRB1*03:01
14 A*29:02g  C*16:01  B*44:03 DRB1*07:01

query_nmdp() returns a list with 4 elements:

1. A data.frame with the original haplotypes
2. A data.frame with the haplotypes found at the nmdp table
3. A data.frame with the haplotypes which were not found at the nmdp table
4. A data.frame with possible haplotypes at nmdp table given the individual's alleles

It is possible to apply query_nmdp() to the whole data, e.g. by using plyr::dlply() with a parallel backend provided by doMC::registerDoMC():

> # using 16 cores:
> doMC::registerDoMC(16)
> 
> results_list <- plyr::dlply(pag, ~subject, . %>% query_nmdp(), .parallel = TRUE)

Using purrr::map_df() it is possible to extract the info for all individuals as a data.frame. For example, let's create a data.frame with all the haplotypes found in the NMDP table:

> haps_found <- 
+   purrr::map_df(results_list, "haplotypes found", .id = "subject") %>%
+   tibble::as_tibble()
> 
> haps_found

# A tibble: 1,352 × 15
   subject        A        C        B        DRB1     AFA_freq AFA_rank
     <chr>    <chr>    <chr>    <chr>       <chr>        <dbl>    <int>
1  HG00096 A*01:01g C*07:01g B*08:01g  DRB1*03:01 1.093801e-02        2
2  HG00096 A*29:02g  C*16:01  B*44:03  DRB1*07:01 3.515995e-03       14
3  HG00097 A*03:01g C*07:02g B*07:02g  DRB1*13:03 8.835382e-05     2047
4  HG00097 A*24:02g C*07:02g B*07:02g  DRB1*15:01 1.471206e-03       54
5  HG00099 A*68:01g C*07:04g B*44:02g  DRB1*03:01 7.084284e-05     2518
6  HG00099 A*01:01g C*07:01g B*08:01g DRB1*11:01g 2.326729e-04      741
7  HG00100 A*01:01g C*07:01g B*08:01g  DRB1*03:01 1.093801e-02        2
8  HG00100 A*01:01g C*06:02g B*57:01g  DRB1*07:01 2.086870e-03       33
9  HG00101  A*32:01 C*02:02g B*27:05g  DRB1*04:04 5.512488e-06    12841
10 HG00101 A*11:01g C*06:02g B*57:01g  DRB1*15:01 0.000000e+00       NA
# ... with 1,342 more rows, and 8 more variables: API_freq <dbl>,
#   API_rank <int>, CAU_freq <dbl>, CAU_rank <int>, HIS_freq <dbl>,
#   HIS_rank <int>, NAM_freq <dbl>, NAM_rank <int>

VitorAguiar/hlahaps documentation built on May 9, 2019, 10:02 p.m.