README.md

In sjmack/BDRR: Calculate Relative Risk for BIGDAWG-formatted Genotype Data

BIGDAWG Relative Risk (BDRR)

version 1.0.3.9000

BDRR includes one function, relRisk(), which calculates relative risk and associated measures for BIGDAWG-formatted non-case-control genotype datasets (shown below), using the fmsb::riskratio function.

Here, “non-case-control” indciates that while two subject categories are required, the dataset should not include patients and controls; instead, the categories may be, e.g., for a dataset of patients, either of two disease states, where one disease state is coded as 0 and the other is coded as 1 in the second column of the dataset.

BIGDAWG::HLA_data[1:4,]

##   SampleID Disease           A           A        DRB1        DRB1        DQB1
## 1  SCo0001       0 01:01:01:01 01:01:01:01    01:01:01    01:01:01 05:03:01:01
## 2  SCo0002       0 03:01:01:01       68:06    08:01:03 15:01:01:01    03:02:12
## 3  SCo0003       0       26:08       32:02 07:01:01:01 15:01:01:01    03:02:01
## 4  SCo0004       0 01:01:01:01    32:01:01    01:01:01    11:04:01 05:03:01:01
##          DQB1  DRB3        DRB3     DRB4  DRB4  DRB5     DRB5
## 1 05:03:01:01 00:00       00:00    00:00 00:00 00:00    00:00
## 2 03:01:01:01 00:00       00:00    00:00 00:00 00:00 01:01:01
## 3 03:01:01:01 00:00       00:00 01:03:03 00:00 00:00 01:01:01
## 4    02:01:01 00:00 02:01:01:01    00:00 00:00 00:00    00:00

Details about the BIGDAWG-format can be found in the BIGDAWG Vignette. BDRR accepts both BIGDAWG-formatted data frames and tab-delimited text files as input.

The riskRatio() function returns an R list object that contains an $alleles and a $genotypes list of analytic result data frames for each locus. Data in these data frames are organized in Locus, Variant, Status_1, Status_0, RelativeRisk, CI.low, CI.high, p.value, and Significant columns, as shown below.

library(BDRR)
exampleRR <- relRisk(BIGDAWG::HLA_data[,1:4])
exampleRR$alleles[[1]][1:5,]    

##   Locus     Variant Status_1 Status_0      RelativeRisk            CI.low
## 1     A 01:01:01:01      176      166  1.03432941327678 0.928482721560761
## 2     A 02:01:01:01       52       58 0.945940890385335 0.774586359500803
## 3     A    02:05:01      105      142 0.843683161448755 0.727283103794645
## 4     A 03:01:01:01      150      145  1.02052995941752 0.908069882145854
## 5     A    03:01:03      137      108  1.12949687674962  1.00601549637306
##             CI.high            p.value Significant
## 1  1.15224258925478  0.545807942758558            
## 2  1.15520258926283  0.576318536881742            
## 3 0.978712791756468 0.0164963281727013           *
## 4  1.14691767511065  0.735265482077059            
## 5  1.26813473469008 0.0518592554276245

exampleRR$genotypes[[1]][1:5,]

##   Locus                 Variant Status_1 Status_0      RelativeRisk
## 1     A 01:01:01:01+01:01:01:01        8        7  1.06936026936027
## 2     A 01:01:01:01+02:01:01:01        4        5 0.890230270511961
## 3     A    01:01:01:01+02:05:01        9       11 0.900910010111224
## 4     A 01:01:01:01+03:01:01:01       13       12  1.04263959390863
## 5     A    01:01:01:01+03:01:03       12        9  1.14691393798899
##              CI.low          CI.high           p.value Significant
## 1 0.664735770005212 1.72027960173905   0.7895603369942            
## 2 0.428256020543676 1.85055176464232 0.742918500771633            
## 3 0.553846329803078 1.46545856249907 0.659670604051562            
## 4 0.713590916343736 1.52341810677183 0.832674625598276            
## 5 0.789817680366197 1.66546231345888 0.504666081455455

sjmack/BDRR documentation built on May 17, 2020, 12:25 a.m.