sharp-match-exposome-proteome.Rmd

---
title: "SHARP data"
author: "ks"
date: "5/10/2021"
output: html_document
---

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

## Exposome

```{r load-data}
load("data/exposome.RData")
ls()
```

```{r data-sizes}
dim(codebook)
dim(covariates)
dim(phenotype)
dim(exposome)
```

```{r covar}
names(covariates)
```


```{r phenotype}
names(phenotype)
```


```{r outcome}
hist(phenotype$hs_zbmi_who)
```


```{r proteome}
load("data/proteome.Rdata")
dim(proteome)
```

Find subset of individuals with both proteome ane exposome.

```{r common-ids}
isctids <- intersect(covariates$ID,sampleNames(proteome))
length(isctids)
```

Filter data to those with proteomic variables.

```{r filter-data}
covars <- covariates %>%
          filter(is.element(ID,isctids))
phenos <- phenotype %>%
          filter(is.element(ID,isctids))
expsms <- exposome %>%
          filter(is.element(ID,isctids))
```

Check order of proteome samples, and match column
order to that in covariate files.

```{r order-proteome-measures}
identical(covars$ID,sampleNames(proteome))
prtm <- exprs(proteome)
prtm <- prtm[,as.character(covars$ID)]
identical(as.character(covars$ID),
          colnames(prtm))
```

```{r cleanup}
rm(proteome)
rm(covariates)
rm(exposome)
rm(phenotype)
rm(isctids)
```

```{r sI}
sessionInfo()
```