Error in making DESeq2 object
1
0
Entering edit mode
2.4 years ago
Tusharika • 0

I have a csv file with the read counts of different samples and genes. I am trying to make a deseq2 object for this csv file but getting the following error.

Command used:

obj_mRNA <- DESeqDataSetFromMatrix(countData = round(database_TCGA_mRNA), colData = clinical_data_TCGA_mRNA, design = ~Sample_Type)

Error:

> Error in h(simpleError(msg, call)) : 
  error in evaluating the argument 'x' in selecting a method for function 'ncol': non-numeric-alike variable(s) in data frame: TCGA-CV-A463-01A, TCGA-CV-7102-01A, TCGA-BA-A6DL-01A, TCGA-CR-6471-01A, TCGA-CV-6961-11A, TCGA-CQ-A4CE-01A, TCGA-CQ-5333-01A, TCGA-DQ-7592-01A, TCGA-UF-A71E-01A, TCGA-D6-A6EM-01A, TCGA-CV-7416-01A, TCGA-CN-4737-01A, TCGA-IQ-A6SH-01A, TCGA-CV-6938-01A, TCGA-CV-6936-11A, TCGA-CR-7401-01A, TCGA-D6-A6EP-01A, TCGA-UF-A71B-01A, TCGA-CR-7402-01A, TCGA-IQ-A61O-01A, TCGA-IQ-7630-01A, TCGA-CQ-A4CB-01A, TCGA-MT-A67F-01A, TCGA-HD-7832-01A, TCGA-QK-A8Z7-01A, TCGA-BA-4075-01A, TCGA-CN-4740-01A, TCGA-H7-8502-01A, TCGA-CV-5966-01A, TCGA-BA-5151-01A, TCGA-CV-7410-01A, TCGA-KU-A6H7-06A, TCGA-CN-5360-01A, TCGA-DQ-5631-01A, TCGA-QK-A6V9-01A, TCGA-CV-5430-01A, TCGA-CN-A497-01A, TCGA-CV-7177-01A, TCGA-CR-7386-01A, TCGA-BA-5153-01A, TCGA-CV-7235-01A, TCGA-BA-5556-01A, TCGA-CN-4742-01A, TCGA-HD-8314-01A, TCGA-HD-7754-01A, TCGA-P3-A6T8-01A, TCGA-CQ-6223-01A, TCGA-CV-5444-01A, TC

What should I do?

Thanks in advance

bioinofrmatics RNASeq deseq2 • 1.4k views
ADD COMMENT
1
Entering edit mode
2.4 years ago
Trivas ★ 1.8k

It looks like you have columns in your data frame called database_TCGA_mRNA that are non-numeric. Try setting those as rownames and trying again.

ADD COMMENT
0
Entering edit mode

I have already set them up as rownames before running the given command.

It looks as follows:

enter image description here

ADD REPLY
0
Entering edit mode

Might be worth creating a second dataframe of your rounded counts and trying to create a DESeq object from this new variable to isolate whether this is an issue with your count matrix or with your colData

ADD REPLY
0
Entering edit mode

I tried rounding the counts but it gives me the same error. I also tried not using the round() function and creating a DESeq2 object but then it gives me another error saying that the count matrix should be numeric. I have attached both the commands and errors.What should I do?

ADD REPLY
0
Entering edit mode

Without having access to your data, I can't really troubleshoot the rest for you. However, both of those lines of code and their subsequent errors show that there are characters in your data frame. I'd recommend subsetting your dataframe (maybe use head and use different numbers of columns) until you can identify where the characters are coming from.

ADD REPLY
0
Entering edit mode

I tried the following code:

type(database)

Output:

"character"

What to do?

ADD REPLY

Login before adding your answer.

Traffic: 2278 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6