error setting up computer for RNA-seq

[gsk7757]

Hello, I have downloaded the scripts according to the computer setup that was mentioned in the RNA-seq by example (sub-book). I have created a new environment "stats" in a new terminal and downloaded below scripts.

Obtain the biostar handbook rnaseq scripts.

curl -O http://data.biostarhandbook.com/books/rnaseq/code.tar.gz

Unpack the code.

tar -xzvf code.tar.gz

Along with the deseq2 and edger packages using mamba mamba install bioconductor-tximport bioconductor-biomart bioconductor-edger bioconductor-deseq2 r-gplots

Followed by: RScript code/parse_featurecounts.r

Running this gave me the following error

Rscript code/combine_transcripts.r [1] "# Tool: Combine transcripts" [1] "# Sample: design.csv" [1] "# Data dir: salmon" Error in file(file, "rt") : cannot open the connection Calls: read.csv -> read.table -> file In addition: Warning message: In file(file, "rt") : cannot open file 'design.csv': No such file or directory Execution halted

same with deseq2

RScript code/deseq2.R Error in file(file, "rt") : cannot open the connection Calls: read.csv -> read.table -> file In addition: Warning message: In file(file, "rt") : cannot open file 'design.csv': No such file or directory Execution halted (stats)

Please help. Thanks.

[ialbert]

well you need a design file of course, follow the tutorials on either the RNA-Seq or the Biostar Workflow pages.

the design lists the groups and the data see the section: What is the experimental design file?

https://www.biostarhandbook.com/books/rnaseq/understand-the-reads.html

[gsk7757]

I followed the biostar workflow of downloading the scripts and ran the simulated data and it worked. However, when I changed the design file to have 4 samples based on the counts file I have. It shows up an error with deseq2 as below: RScript code/deseq2.R Error in $<-.data.frame(*tmp*, condition, value = integer(0)) : replacement has 0 rows, data has 10 Calls: $<- -> $<-.data.frame Execution halted (stats)

[ialbert]

you have to show the design and the first few lines of the count,

the two have to match for this work correctly

[gsk7757]

could you please direct me to the workflow that shows how to edit the Rscript to make changes to the design file and count file?

[ialbert]

you don't need to edit the count file, the design file is a simple text file you can edit it with any editor,

the point I was making is that the design file has to match the count file. if the columns are named differently than in the count file then it cannot match the design to the counts

[gsk7757]

Sorry, my bad. I have named the columns to match the count file. But still i see the same error. Below is my design

sample	condition
C1dedup	Control
C2dedup	Control
E1dedup	Experimental
E2dedup	Experimental

And this is how my counts file look like

Geneid	C1dedup	C2dedup	E1dedup	E2dedup
CAF26817-1	551	545	323	208
CAF26818-1	219	212	103	68
CAF26819-1	547	397	106	81

[ialbert]

please paste the actual file (the complete design and first few lines of the counts ) not a markdown formatted version of it.

Since you are formatting the file by hand, it cannot possibly be the actual file right?

[gsk7757]

This is my actual counts file

# Program:featureCounts v2.0.6; Command:"featurecounts" "-a" "Bartonella_henselae_str_houston_1_gca_000046705.ASM4670v1.57.gff3" "-g" "Name" "-o" "counts.txt" "C1dedup.bam" "C2dedup.bam" "E1dedup.bam" "E2dedup.bam"                                  
Geneid  Chr Start   End Strand  Length  C1dedup C2dedup E1dedup E2dedup
CAF26817-1  Chromosome  1   825 +   825 551 545 323 208
CAF26818-1  Chromosome  837 1433    +   597 219 212 103 68
CAF26819-1  Chromosome  1426    2307    +   882 547 397 106 81
CAF26820-1  Chromosome  2304    2891    +   588 375 354 458 278
CAF26821-1  Chromosome  2896    3603    +   708 143 126 146 119
CAF26822-1  Chromosome  3932    6838    +   2907    1442    1444    600 373
CAF26823-1  Chromosome  6854    7735    -   882 917 861 113 68
CAF26824-1  Chromosome  7827    8600    -   774 22982   26927   3981    2544
CAF26825-1  Chromosome  9215    11056   -   1842    741 598 371 216
CAF26826-1  Chromosome  11591   12082   -   492 2674    2637    664 355
CAF26827-1  Chromosome  12079   12942   -   864 388 314 197 132
CAF26828-1  Chromosome  12930   13862   -   933 483 341 149 114
CAF26829-1  Chromosome  14030   14326   -   297 387 442 282 167
CAF26830-1  Chromosome  14335   14613   -   279 868 974 434 286
CAF26831-1  Chromosome  14805   15455   +   651 4480    4431    776 480
CAF26832-1  Chromosome  15567   16145   +   579 2314    2312    376 228

From this, I made another counts.csv file by just taking columns 1, 7-10. which looks like below

Geneid	C1dedup	C2dedup	E1dedup	E2dedup
CAF26817-1	551	545	323	208
CAF26818-1	219	212	103	68
CAF26819-1	547	397	106	81
CAF26820-1	375	354	458	278
CAF26821-1	143	126	146	119
CAF26822-1	1442	1444	600	373
CAF26823-1	917	861	113	68
CAF26824-1	22982	26927	3981	2544
CAF26825-1	741	598	371	216
CAF26826-1	2674	2637	664	355
CAF26827-1	388	314	197	132
CAF26828-1	483	341	149	114
CAF26829-1	387	442	282	167
CAF26830-1	868	974	434	286
CAF26831-1	4480	4431	776	480
CAF26832-1	2314	2312	376	228

and I made design file that looks as below

sample	condition
C1dedup	Control
C2dedup	Control
E1dedup	Experimental
E2dedup	Experimental

I closed the terminal and reopened it and ran deseq2 again, which gave me a new error. RScript code/deseq2.R Warning message: In read.table(file = file, header = header, sep = sep, quote = quote, : incomplete final line found by readTableHeader on 'design.csv' Error in DESeqDataSetFromMatrix(countData = countData, colData = colData, : ncol(countData) == nrow(colData) is not TRUE Calls: DESeqDataSetFromMatrix -> stopifnot Execution halted

v

[ialbert]

Again the files that you include as a CSV file do not look like are comma-separated files. The file should look like this

a,b,c
1,2,3

instead, the content you show looks like this

a	b	c
1	2	3

do you see the difference? The design files you show do not look like comma-separated files.

[gsk7757]

Sorry, I was opening them in an Excel sheet and copy-pasting them. When I open the files in the terminal this is how they look. more design.csv

sample,condition C1dedup,Control C2dedup,Control E1dedup,Experimental E2dedup,Experimental (stats) and counts more counts.csv Geneid,C1dedup,C2dedup,E1dedup,E2dedup CAF26817-1,551,545,323,208 CAF26818-1,219,212,103,68 CAF26819-1,547,397,106,81 CAF26820-1,375,354,458,278 CAF26821-1,143,126,146,119 CAF26822-1,1442,1444,600,373 CAF26823-1,917,861,113,68 CAF26824-1,22982,26927,3981,2544 CAF26825-1,741,598,371,216 CAF26826-1,2674,2637,664,355

[gsk7757]

Finally, it worked. I had to rerun codes RScript code/parse_featurecounts.r RScript code/deseq2.r When i ran the codes the first time, somehow the output of the counts.csv file was weird. But it worked the second time.

Thank you so much for the help.