Normalize HLA typing output of Seq2HLA, Optitype & ATHLATES.
Hlarp
is also provided as a library.
Pronounced "heh-larp" (or with a soft 'h' : "larp").
The easiest way to install the tool is to "opam pin" this repository:
opam pin add -k git https://github.com/hammerlab/hlarp
Now you can run hlarp with hlarp
.
hlarp seq2HLA /path/to/seq2HLA/results directory > results.csv
hlarp optitype /path/to/OptiType/results directory > results.csv
The output is a csv
class | allele | qualifier | confidence | run |
---|---|---|---|---|
1 | A*02:05:01 | 0.500000 | 120013 | |
1 | A*30:01:01 | 0.500000 | 120013 | |
1 | B*57:03:01 | 1.000000 | 120013 | |
1 | B*57:03:01 | 1.000000 | 120013 | |
1 | C*18:02 | 1.000000 | 120013 | |
1 | C*18:02 | 1.000000 | 120013 | |
2 | DRB1*03:02:01 | 0.500000 | 120013 | |
2 | DRB1*15:03:01:01 | 0.250000 | 120013 | |
2 | DRB3*01:01:02:01 | 0.500000 | 120013 | |
2 | DRB3*01:01:02:02 | 0.500000 | 120013 | |
2 | DRB5*01:01:01 | 1.000000 | 120013 | |
2 | DRB5*01:01:01 | 1.000000 | 120013 | |
1 | A*31:01:02 | 0.500000 | 120021 | |
1 | A*31:01:13 | 0.500000 | 120021 | |
1 | C*02:10 | 1.000000 | 120021 | |
1 | C*02:10 | 1.000000 | 120021 | |
2 | DRB1*15:03:01:01 | 0.500000 | 120021 | |
2 | DRB1*15:03:01:02 | 0.500000 | 120021 | |
2 | DRB3*02:02:01:01 | 0.500000 | 120021 | |
2 | DRB3*02:02:01:02 | 0.500000 | 120021 | |
2 | DRB5*01:01:01 | 1.000000 | 120021 | |
2 | DRB5*01:01:01 | 1.000000 | 120021 | |
1 | A*68:02:01:01 | 0.166667 | 120074 | |
1 | A*74:01 | 0.500000 | 120074 | |
1 | B*15:03:01 | 0.500000 | 120074 | |
1 | B*15:16:01 | 0.500000 | 120074 | |
1 | C*02:10 | 0.500000 | 120074 | |
1 | C*16:01:01 | 0.500000 | 120074 | |
2 | DRB1*01:02:01 | 0.500000 | 120074 | |
2 | DRB1*14:54:01 | 0.500000 | 120074 |
Some columns are left empty due to the nature of the HLA-typer.
hlarp compare --resolution 2 -l A -l B -l C --loci DRB1 --optitype /path/to/optitype/results -a /path/to/ATHLATES/results/
Will generate this kind of report:
120013 jacard similarities: 1.00 1.00 0.00 0.00
A: A*02:05 OptiType;ATHLATES
A*30:01 OptiType;ATHLATES
B: B*57:03 OptiTypex2;ATHLATESx2
C: C*18:01 OptiTypex2
C*18:02 ATHLATESx2
DRB1: DRB1*03:02 ATHLATES
DRB1*15:03 ATHLATES
120021 jacard similarities: 0.33 0.00 0.33 0.00
A: A*02:01 OptiType
A*31:01 OptiType;ATHLATESx2
B: B*15:03 OptiType
B*45:01 OptiType
C: C*02:10 OptiType;ATHLATESx2
C*16:01 OptiType
DRB1: DRB1*15:03 ATHLATESx2
120074 jacard similarities: 1.00 1.00 1.00 0.00
A: A*68:02 OptiType;ATHLATES
A*74:01 OptiType;ATHLATES
B: B*15:03 OptiType;ATHLATES
B*15:16 OptiType;ATHLATES
C: C*02:10 OptiType;ATHLATES
C*16:01 OptiType;ATHLATES
DRB1: DRB1*01:02 ATHLATES
DRB1*14:54 ATHLATES
...
Average jacard similarities across runs: 0.54 0.60 0.51 0.07
You can also use the hlarp
module:
open Hlarp