Question

Filtering output from tximport

0

Entering edit mode

Sheean • 0

@a1d6aa85

Last seen 3.0 years ago

United States

Hello,

I would like to filter the resulting txi file generated from tximport of Salmon files to exclude columns where there are >=4 0s in 'counts'.

I did: txi <-tximport(files, type = "salmon", txOut = TRUE)

this yields txi, a large list of 4 elements, where:

attributes(txi) $names [1] "abundance" "counts" "length" "countsFromAbundance"

I want to remove/exclude columns where there are >=4 0s in "counts." Counts are in columns 22-41.

To do this, I did:

txi_subset = txi[!rowSums(txi[, c(22:41)] == 0) >=4, ]

I get:

Error in h(simpleError(msg, call)) : error in evaluating the argument 'x' in selecting a method for function 'rowSums': incorrect number of dimensions

Any help would be appreciated. Thank you.

tximport Salmon • 1.9k views

ADD COMMENT • link updated 3.0 years ago by Michael Love 43k • written 3.0 years ago by Sheean • 0

score 0 · Answer 1 · 2021-11-19

0

Entering edit mode

James W. MacDonald 67k

@james-w-macdonald-5106

Last seen 1 day ago

United States

You already acknowledged that your 'txi' object is a list, but then you tried to subset it as if it were a data.frame or a matrix.

ADD COMMENT • link 3.0 years ago James W. MacDonald 67k

0

Entering edit mode

Right. Recommend tximeta package and function which will give a SummarizedExperiment.

Most users who want a Bioc-like object benefit from tximeta over tximport, the former calls the latter anyway.

coldata <- data.frame(
  files=c("/path/sample1/quant.sf",...), 
  names=c("s1",...), 
  condition=...
)
se <- tximeta(coldata)

ADD REPLY • link 3.0 years ago Michael Love 43k

0

Entering edit mode

Thanks, I'm not totally sure this will work to resolve the need to remove 0's. Regardless, I'm getting:

Error: lexical error: invalid char in json text. ./cmd_info.json (right here) ------^

I restarted R Studio and got the same error.

ADD REPLY • link 3.0 years ago Sheean • 0

0

Entering edit mode

That error has been resolved by previous users here by using up to date package versions.

ADD REPLY • link 3.0 years ago Michael Love 43k