fittingtool_notebook.Rmd

---
title: "Fitting tool notebook"
output: html_notebook
---

Load libraries and util

```{r}
library('NMF')
library("ggplot2")
library('MutationalPatterns')
source(file="utils.R")
```

Load data
```{r}
breast_cancer = read.csv(file="data/breast_cancer.csv",header=TRUE)
cosmic_signatures = read.csv(file="data/cosmic_signatures.csv",header=TRUE)
cosmic_signatures=cosmic_signatures[,3:ncol(cosmic_signatures)]
BC_exposure= read.delim(file = "data/extanded_breast_cancer_expousre.txt",header=T,sep="\t")
noise_BC_samples= read.delim(file = "data/noise_breast_cancer_samples.txt",header=T)
```

```{r}
as.matrix(BC_exposure)
nrow(BC_exposure)
ncol(BC_exposure)
rownames(BC_exposure)
```


```{r}
strsplit(colnames(BC_exposure)[1],"\\.")[[1]]

toString(BC_exposure[1,])

strsplit(toString(BC_exposure[1,])," ")[[1]]
```

Breast cancer exposure
```{r}
BC_exposure= read.delim(file = "data/extanded_breast_cancer_expousre.txt",header=T,sep="\t")

BC_expo=matrix(0, nrow=560, ncol=66)

colnames(BC_expo)=strsplit(colnames(BC_exposure)[1],"\\.")[[1]]

for(i in seq(1,560)){
  spl=strsplit(toString(BC_exposure[i,])," ")[[1]]
  for(j in seq(1,66)){
    BC_expo[i,j]=as.numeric(spl[j+1])
  }
}

```

Breast cancer samples (ideal + noise)
```{r}
BC_samples=matrix(0, nrow=96, ncol=560)

colnames(BC_samples)=strsplit(colnames(noise_BC_samples)[1],"\\.")[[1]]

for(i in seq(1,96)){
  spl=strsplit(toString(noise_BC_samples[i,])," ")[[1]]
  for(j in seq(1,560)){
    BC_samples[i,j]=as.numeric(spl[j+1])
  }
}
```

```{r}
source(file="add_signatures.R")
source(file="fitting_probabilities.R")
#source(file="Hparams.R")
#source(file="main.R")
source(file="probabilities_to_exposure.R")
source(file="remove_signatures.R")
source(file="remove_small_exposure.R")
source(file="remove_specific_signatures.R")
source(file="sparse_filter.R")
```

```{r}
probs=fitting_probabilities(mut_mat=BC_samples[,1:5],
                      signatures = cosmic_signatures,
                      ind_prior_knowledge = c(1))
```