bcl-convert cannot create output directory structure at the path specified

Question

[giulia.trauzzi]

Hi all,

I am running bcl-convert for the first time as a customer is interested in getting the fastq files for his index reads without demultiplexing the data - which to me, does not sound possible or useful. I am not finding much about this online so I was hoping someone could enlighten me on this whether it is possible (anyway I am doing some tests).

My main issue at the moment is that I am running bcl-convert and whatever path I specify as output-directory, my job keeps failing with the error message being

ERROR: Output folder directory structure can not be created at /data09/bcl2fast

Has anyone had the same issue? How can this be solved?

Thanks,

Giulia

[marco.barr]

I've never received this error before.. but you could try checking the permissions

[giulia.trauzzi]

Hi, thanks for this. I checked the permissions of the folder and changed them with

chmod ugo+rwx /output_directory

This has changed indeed the permissions but I still get the same error message.

Any idea?

Best, Giulia

[GenoMax]

Is that the only error? Can you provide the entire command line you are using?

as a customer is interested in getting the fastq files for his index reads without demultiplexing the data - which to me, does not sound possible or useful.

There are applications that expect the index reads to be provided in a separate file so this is not out of ordinary.

[giulia.trauzzi]

Thanks! I am sure there are. Are these necessary to demultiplex the data though? I am assuming most demultiplexing tools do not require these files.

This is the command line I am using. Yes, at the moment, that is the only error I am getting and cannot get past it.

bcl-convert --bcl-input-directory ../240521_M03762_0256_000000000-DNBGM --force --output-directory ./fastq/

Giulia

[GenoMax]

Looks like you are not using a samplesheet.

You will have to use a samplesheet though if you want to create fastq files for index reads since that directive goes into the SampleSheet file: https://knowledge.illumina.com/software/on-premises-software/software-on-premises-software-faq-list/000007493

[BCLConvert_Settings]

CreateFastqForIndexReads,1

[giulia.trauzzi]

Hi, I am using a SampleSheet. And I am also using the settings to create the reads for the index, but I am a bit confused as to what the samplesheet should look like for bcl convert as opposed to bcl2fastq (which I usually use). After having a look, I came up with this samplesheet

[Header],,,,,,
IEMFileVersion,5,,,,,
Instrument Type,MiSeq,,,,,
[Reads],,,,,,
251,,,,,,
251,,,,,,
[Settings],,,,,, #should this be [BCLConvert_Settings] ?
CreateFastqForIndexReads,1,,,,, #to create the fastq for the reads
OverrideCycles,Y251;N6;N8;Y251,,,,, #to skip demultiplexing
[Data],,,,,, #should this be [BCLConvert_Data] ?
Sample_ID,Sample_Plate,Sample_Well,I7_Index_ID,index,I5_Index_ID,index2
1,,,ILL_R01_P1_AV09,CGTGAT,ILL_F01_P1_AV09,GAAGAGAT
#etc other samples

When listing all the samples and index sequences, should I just write down my normal samples and their index sequences or is there a specific format that should go with those specific settings?

Because when I do this, I get another error

ERROR: Sample Sheet Error: More than one sample specified for lane 1 on a no-index run
Assertion failed in ../src/host/dragen_api/file_io/bcl/sample_sheet.cpp line 976 -- false --

I am struggling to find an example of samplesheet to use with BCLConvert.

PS. Tech support solved my first error message, saying that my working directory was not seen by the singularity container that is why I could not "recreate the folder structure in my wdir.

Thanks, Giulia

[GenoMax]

If you are able to use a windows program then download Illumina Experiment Manager to create the samplesheet.

was not seen by the singularity container

Guess you had left that critical bit of information out in the original post :-)

[vikasmh111]

Hello, Were you able to solve the problem? I am facing the same issue.

Thank you

[GenoMax]

Every case can be different. Just saying "facing the same issue" is not enough information. Tell us how you are using the software (directly, via a container etc) and the command line you are using.

[giulia.trauzzi]

Hi, I think I did in the end but just doing something completely different in a way that I ended up using bcl2fastq specifying --create-fastq-for-index-reads) and the samplesheet looked like this

[Header],,,,,, 
IEMFileVersion,5,,,,,
Experiment Name,
[Reads],,,,,,
251,,,,,,
251,,,,,,
[Settings],,,,,,
CreateFastqForIndexReads,1,,,,,
OverrideCycles,Y251;N6;N8;Y251,,,,,
[Data],,,,,,
Sample_ID,Sample_Plate,Sample_Well,I7_Index_ID,index,I5_Index_ID,index2
All_samples,,,ILL_R01_P1_AV09,CGTGAT,ILL_F01_P1_AV09,GAAGAGAT #basically renamed all samples the same so they would be outputted to the same file 
All_samples,,,ILL_R05_P1_AV09,CACTGT,ILL_F02_P1_AV09,CCAGCTTG

This has allowed me to generate fastq files for index sequences without demultiplexing. This was a flexible way of doing it as Illumina support could not help and I could not find another way.

Hope this helps.

Giulia