How to download RNAseq data of only male KIRC patients?

Question

[bioinfo_enthusiast]

Hello everyone. I am want to perform WGCNA analysis of only male Kidney Clear Cell Carcinoma (KIRC) patients. I am trying to download trancriptome data of male KIRC patients from TCGA using tcgabiolinks package. Here is my code:

query_TCGA <- GDCquery(project = 'TCGA-KICH',
                       data.category = 'Transcriptome Profiling',
                       experimental.strategy = 'RNA-Seq',
                       workflow.type = 'STAR - Counts',
                       access = "open",
                       sample.type = "Primary Tumor",
                       )

However, I am not able to specify the gender = "Male", and disease = "KIRC". How can I download my desired data?

Thank you so much.

Answer 1

[joe]

The gender filter is simple, the KIRC 'Kidney renal clear cell carcinoma' might be more difficult. Are you expecting to find a label as KIRC?

library(TCGAbiolinks)
library(DESeq2)
library(data.table)

query_TCGA <- GDCquery(project = 'TCGA-KICH',
                       data.category = 'Transcriptome Profiling',
                       experimental.strategy = 'RNA-Seq',
                       workflow.type = 'STAR - Counts',
                       access = "open",
                       sample.type = "Primary Tumor")

GDCdownload(query = query_TCGA)
data <- GDCprepare(query = query_TCGA)

data.dt <- data.table(
  as.data.frame(colData(data)), 
  options = list(scrollX = TRUE, keys = TRUE, pageLength = 5), 
  rownames = FALSE
)

data.dt[data.dt$gender=="male" & data.dt$primary_diagnosis=="Renal cell carcinoma, chromophobe type"]

[joe]

Just to expand on that a little, you can search for text in the metadata with the below - I couldn't find "KIRC" but could find "male" ....so likely not there

grep("male", capture.output(str(data)))
grep("KIRC", capture.output(str(data)))

[bioinfo_enthusiast]

Thank you so much.

I have a question. I am new in the field of bioinformatics. Would you kindly tell me whether my study design is correct or not? I am using only primary tumor data of male KIRC patients to perform WGCNA. I'm not including solid tissue data, which is normally control data. Does it needed to have both case and control for WGCNA? Or I can only perform on case data?

[joe]

If you're new to bioinformatics, I recommend to explore the vignettes published by the developers of TCGAbiolinks and WGCNA. I would expect that the control is necessary, but maybe depending on precisely what you want to learn you can get away without using one. Generally, controls in bioinformatics are required, and if the data already exists you should definitely include (or at least process with and without and check what changes). Also broadly speaking, you should conduct a power analysis to see if the data you have available can even offer meaningful results.

[bioinfo_enthusiast]

Okay. Thank you so much!!