dip_to_hap.Rmd

---
title: "Haploidizing diploid data"
author: "Brian J. Knaus"
date: "February 7, 2017"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```


Load example data set and extract the genotypes.


```{r}
library(vcfR)
data(vcfR_test)
head(vcfR_test)
gt <- extract.gt(vcfR_test)
head(gt)
```


Use `is_het()` to identify heterozygous positions and censor as NA.


```{r}
is.na(vcfR_test@gt[,-1][is_het(gt)]) <- TRUE
vcfR_test
```


Extract the remaining genotypes and use `unique()` to query for the unique genotypes.


```{r}
gt <- extract.gt(vcfR_test)
unique(as.vector(gt))
```


We can now manually set each genotype to a haploid state.
Note that your genotypes will be different!


```{r}
gt[gt=="0|0"] <- 0
gt[gt=="0/0"] <- 0
gt[gt=="1/1"] <- 1
gt[gt=="2/2"] <- 2
head(gt)
```


Now extract everything but the genotype from the vcfR object, paste our haploid genotypes to it and update the vcfR object.


```{r}
gt2 <- extract.gt(vcfR_test, extract = FALSE)
head(gt2)
gt <- matrix( paste(gt, gt2, sep=":"), nrow=nrow(gt), dimnames=dimnames(gt) )
is.na(gt[gt == "NA:NA"]) <- TRUE
vcfR_test@gt[,-1] <- gt
head(vcfR_test)
```