---
title: "Working with simple somatic mutations"
author: "Sean Davis"
date: "`r format(Sys.Date(), '%A, %B %d, %Y')`"
always_allow_html: yes
output:
  BiocStyle::html_document:
    df_print: paged
    toc_float: true
abstract: >

vignette: >
  %\VignetteIndexEntry{Somatic Mutation Data}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

# Background



# Workflow

```{r warning=FALSE,message=FALSE}
library(GenomicDataCommons)
library(tibble)
```

## Genes and gene details

```{r}
grep_fields('genes', 'symbol')
```
```{r}
head(available_values('genes','symbol'))
```



```{r}
tp53 = genes() %>% 
  GenomicDataCommons::filter(symbol=='TP53') %>% 
  results(size=10000) %>% 
  as_tibble()
```


## ssms

```{r}
ssms() %>% 
    GenomicDataCommons::filter(
      chromosome==paste0('chr',tp53$gene_chromosome[1]) &
        start_position > tp53$gene_start[1] & 
        end_position < tp53$gene_end[1]) %>% 
    GenomicDataCommons::count()
```

```{r}
ssms() %>% 
    GenomicDataCommons::filter(
      consequence.transcript.gene.symbol %in% c('TP53')) %>% 
    GenomicDataCommons::count()
```

## convert to VRanges

```{r warning=FALSE,message=FALSE}
library(VariantAnnotation)
vars = ssms() %>% 
    GenomicDataCommons::filter(
      consequence.transcript.gene.symbol %in% c('TP53')) %>% 
    GenomicDataCommons::results_all() %>%
    as_tibble()
```

```{r}
vr = VRanges(seqnames = vars$chromosome,
             ranges = IRanges(start=vars$start_position, width=1),
             ref = vars$reference_allele,
             alt = vars$tumor_allele)
```

```{r}
ssm_occurrences() %>% 
    GenomicDataCommons::filter(
      ssm.consequence.transcript.gene.symbol %in% c('TP53')) %>% 
    GenomicDataCommons::count()
```

```{r}
var_samples = ssm_occurrences() %>% 
    GenomicDataCommons::filter(
      ssm.consequence.transcript.gene.symbol %in% c('TP53')) %>% 
    GenomicDataCommons::expand(c('case', 'ssm', 'case.project')) %>%
    GenomicDataCommons::results_all() %>% 
    as_tibble()
```

```{r}
table(var_samples$case$disease_type)
```

## OncoPrint

```{r}
fnames = files() %>%
  GenomicDataCommons::filter(
    cases.project.project_id=='TCGA-SKCM' & data_format=='MAF' & data_type=='Masked Somatic Mutation' & analysis.workflow_type=='MuTect2 Variant Aggregation and Masking') %>%
  ids() %>%
  gdcdata()
```

```{r cache=TRUE}
library(maftools)
melanoma = read.maf(maf = fnames[1])
```

```{r}
maftools::oncoplot(melanoma)
```