lwaldron

---
title: "Anscombe residuals plots"
author: "Levi Waldron"
date: "`r Sys.Date()`"
output: html_document
---

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

lwaldron

# installed curatedMetagenomicData from github
suppressPackageStartupMessages({
    library(curatedMetagenomicData)
    library(dplyr)
})

#I download the data in two ways, one select a few CRC studies and the other will multiple studies available.

# specific only crc studies downloaded
suppressMessages({

lwaldron

# See https://diabimmune.broadinstitute.org/diabimmune/antibiotics-cohort/resources/16s-sequence-data
# The provided command `wget -r -np -nd https://pubs.broadinstitute.org/diabimmune/data/15` does not work because files are listed in an html page
library(dplyr)
library(rvest)
url <- "https://diabimmune.broadinstitute.org/diabimmune/data/15/"
url %>%
  read_html() %>%
  html_elements("a") %>%
  html_attr("href") %>%
  download.file(., destfile = basename(.))

lwaldron

##### Importing COVID-19 data from the NYC DOHMH github (https://github.com/nychealth/coronavirus-data) and merge with ACS data of interest

# In order to get the URL of a table of your interest, go to the table and click on  'History' on the top right corner. 
# You will see the upload history for the table on this page. Choose a time point of interest and click on the second 
# to the last button on the right (if you hover over the button it should say 'View at this point in the hisotry'). 
# You will be directed to view the table. Then click on 'Raw' and copy the URL. 

covid <- read.csv("https://raw.githubusercontent.com/nychealth/coronavirus-data/7ce1b84610232be9c3f780484865a51f73b8c469/recent/recent-4-week-by-modzcta.csv")
head(covid)

lwaldron

##### Importing Framingham Heart Study data from a github repository (https://github.com/GauravPadawe/Framingham-Heart-Study) 

library(tidyverse)

#importing the dataset
chddata  <- read.csv("https://raw.githubusercontent.com/GauravPadawe/Framingham-Heart-Study/adcc828b8a5b3ddbd8d5b8b98e2b27cf60538db6/framingham.csv")

#some recoding
chddataclean <- chddata %>%
  mutate(TenYearCHD = if_else (TenYearCHD=='1',"CHD", "No-CHD"),

lwaldron

library(SingleCellMultiModal)
library(MultiAssayExperiment)

suppressMessages(scmm <- scMultiome(dry.run = FALSE))
format(object.size(scmm), units="Mb") #31Mb in memory
saveHDF5MultiAssayExperiment(scmm)
dir("h5_mae", full.names=TRUE) |> file.info() # ~193MB on disk


suppressMessages(scmm_sparse <- scMultiome(format = "MTX", dry.run = FALSE))

lwaldron

## ---------------------------------------------------------------------------------------------------------------------------------
library(curatedTCGAData)
library(TCGAutils)
library(RaggedExperiment)


## -----------------------------------------------------------------------------------------------------------------------------------------
cnvdry <-
  curatedTCGAData(assays = "CNVSNP",

lwaldron

# I create and discuss this code at https://youtu.be/nU_GEPKVXU8

library(dplyr)
#> 
#> Attaching package: 'dplyr'
#> The following objects are masked from 'package:stats':
#> 
#>     filter, lag
#> The following objects are masked from 'package:base':
#> 

lwaldron

system("gsutil cp gs://biocbbs_2020a/cmd3out/uploads/GuptaA_2019.metaphlan_bugs_list.stool.rda .")
load("GuptaA_2019.metaphlan_bugs_list.stool.rda")
head(rownames(GuptaA_2019.metaphlan_bugs_list.stool)) #first 3 look wrong
grep("CIBIO", rownames(GuptaA_2019.metaphlan_bugs_list.stool)) #there are 60 with CIBIO in the rowname

lwaldron

# to run this using Docker from the command line on the stock Bioconductor image:
# docker run -it bioconductor/bioconductor_docker:latest R

BiocManager::install("cBioPortalData")
library(cBioPortalData)

#acc_tcga full data pack
system.time(accpack <- cBioDataPack("acc_tcga")) #~10 seconds
accpack