Possible data corruption?

[concatenize]

Hi Dr. Veres,
I'm working with some of your data, downloaded from GEO. When I read into R the file x1_S3c_b1.counts.tsv.gz from Veres2019_analysis_data⁩ ▸ ⁨01_Stages_3_to_6⁩ ▸ ⁨data⁩ ▸ ⁨indrops_raw⁩, I get the classic R error Error in scan(file, what, nmax, sep, dec, quote, skip, nlines, na.strings, : line 14347 did not have 41748 elements. I'm reading it like this:

fname = "x1_S3c_b1.counts.tsv.gz"
expr = read.table(file.path("data/melton_veres/01_Stages_3_to_6", fname) %>% gzfile, skip = 14346, nrows = 3 )

I looked for the offending line, comparing it to the ones before it like this:

scan(file.path("data/melton_veres/01_Stages_3_to_6", fname) %>% gzfile, skip = 14342, nlines = 3, what = "character" ) %>% length %>% divide_by(3)
> 41748
scan(file.path("data/melton_veres/01_Stages_3_to_6", fname) %>% gzfile, skip = 14346, nlines = 3, what = "character" ) %>% length %>% divide_by(3)
> 82838290

It looks to me like I am reading a corrupted file with a "frameshift mutation" that wipes out all the newlines after line 14346. Could you check and see if the file looks OK on your end?

Thanks!