---
title: "tLOH"
author: "Michelle Webb"
date: "5/5/2021"
output: html_document
vignette: >
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteIndexEntry{tLOH}
  %\usepackage[UTF-8]{inputenc}
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
library('tLOH')
library('purrr')
library('dplyr')
library('GenomicRanges')
library('VariantAnnotation')
```
v0.99.10

tLOH: Assessment of evidence for loss of heterozygosity in spatial 
transcriptomics pre-processed data using Bayes factor calculations.

This tool requires data produced with the 10X Genomics Visium 
Spatial Gene Expression platform and processed to obtain a VCF with per-cluster 
allele count information at heterozygous SNP positions. The purpose of this R 
package is to perform Bayes calculations on the data from the VCF and plot the 
results. Examples of how to run this tool are below:


1. tLOHDataImport() is a function to import data for use by tLOHCalc(). The 
input is a VCF file, an example can be found in the inst/extdata folder.

```{r dataImport, eval=FALSE, warnings = FALSE}
exampleData <- tLOHDataImport('../inst/extdata/Example.vcf')
# The VCF file in inst/extdata must be decompressed before running this command
```

2. tLOHCalc() is the main calculation function. It requires the output 
directory from tLOHDataImport().
```{r marginalCalculations, eval=TRUE}
load("../data/humanGBMsampleAC.rda")
df <- tLOHCalc(humanGBMsampleAC)
head(df)
```

The column descriptions for the output dataframe are as follows -
CHR: chromosome
POS: position
REF: reference allele counts
ALT: alternative allele counts
TOTAL: total counts
p(D|het): probability of data given heterozygous event
p(D|loh): probability of data given loh event
p(het|D): probability of data given heterozygous event divided by the addition 
of p(D|het) and p(D|loh)
p(loh|D): probability of data given loh event divided by the addition of 
p(D|het) and p(D|loh)
bayesFactors: Bayes factor value K
inverseBayes: 1/K
LogBayesFactors: log of Bayes factor K
LogInverseBayes: log of 1/K 
Log10BayesFactors: log 10 of Bayes factor K
Log10InverseBayes: log 10 of 1/K
AF: allele fraction
Cluster: cluster number
Cluster_AF: cluster + AF for plotting y axis
CHR_F: chromosome factor

3. tLOH has two plotting functions, alleleFrequencyPlot() and 
aggregateCHRPlot(). Both require the output dataframe from running tLOHCalc, 
and a sample name for the plot title. Example images of these plots are 
available in inst/extdata.

```{r plotting, eval=TRUE}
alleleFrequencyPlot(df, "Example")

aggregateCHRPlot(df, "Example")
```