bcl2fastq v2.19.1.403 : error during demultimpexing single and dual index reads

Question

[sourav.por]

Hi,

I have a HiSeq2500 run data. Out of its 8 lanes, lane 1 contains single six index length and the rest of all the seven lanes contains dual eight indexes.

Below is the RunInfo file, for the run

RunInfo Version="2"
Run Id="180524_D00239_0216_Acb0yeanxx" Number="216"
Blockquote
Flowcell cb0yeanxx
Instrument D00239
Date 180524
Reads
Read Number="1" NumCycles="125" IsIndexedRead="N"
Read Number="2" NumCycles="8" IsIndexedRead="Y"
Read Number="3" NumCycles="8" IsIndexedRead="Y"
Read Number="4" NumCycles="125" IsIndexedRead="N"
Reads
FlowcellLayout LaneCount="8" SurfaceCount="2" SwathCount="3" TileCount="16"
AlignToPhiX
Run
RunInfo

I also have my Sample sheet in the below format.

Instrument Type,    HiSeq 2500
Assay,  paired read 250 cycles
Index Adapters, TrueSeq dual index adapters 96 for illumina
Description,    XXX-059_human_exom
Chemistry,  v4 dual index
[Reads]
125
125
[Data]
Lane,Sample_ID,Sample_Name,Plate,Well,Index Well,I7_Index_ID,index,I5_Index_ID,index,Sample_Project,Description
1,YYY3401,YYY3401,,,D7,TAATGCGC,,,,
1,YYY3407,YYY3407,,,D7,TAATGCGC,,,,
2,ZZZ3409,ZZZ3409,,,D7,TAATGCGC,D5,CCTATCCT,,
2,ZZZ3411,ZZZ3411,,,D7,TAATGCGC,D5,GGCTCTGA,,
2,ZZZ3410,ZZZ3410,,,D7,CGGCTATG,D5,CCTATCCT,,

I used the below command for demultiplexing,

bcl2fastq --runfolder-dir ../../180524_D00239_0216_Acb0yeanxx/ --create-fastq-for-index-reads --use-bases-mask 1:Y125,I6,I*,Y125 --use-bases-mask Y125,I8,I8,Y125 --sample-sheet ../SampleSheet_for_Exom_24_05_18.csv -p 10 --output-dir ../FASTQ_SSB_059/

and I am getting the below error,

ERROR: bcl2fastq::common::Exception: 2018-Jun-02 13:28:25: Success (0): /TeamCityBuildAgent/work/556afd631a5b66d8/src/cxx/lib/layout/UseBasesMask.cpp(219): Throw in function void bcl2fastq::layout::UseBasesMask::parseAsteriskAndValidateSize(std::string&, bool&, unsigned int) const
Dynamic exception type: boost::exception_detail::clone_impl<bcl2fastq::layout::UseBasesMaskFormatError>
std::exception::what: One or more read masks do not match the number of cycles specified in RunInfo.xml, and a '*' was specified. There is no way to determine the number of cycles implied by the '*'. Base mask: i*'

Can anyone please help me regarding this issue?

[zhengpanone]

I have the same problem. How do you set --use-bases-mask parameters?

[zhengpanone]

ERROR: bcl2fastq::common::Exception: 2018-Sep-22 13:20:09: Success (0):/TeamCityBuildAgent/work/556afd631a5b66d8/src/cxx/lib/layout/UseBas    esMask.cpp(181): Throw in function void bcl2fastq::layout::UseBasesMask::expandNumbers(std::string&) const
 Dynamic exception type: boost::exception_detail::clone_impl<bcl2fastq::layout::UseBasesMaskFormatError>
 std::exception::what: UseBasesMask formatting error. Error parsing number from mask. Base mask: 'i0n8'

[GenoMax]

You set the --use-bases-mask as follows:

--use-bases-mask Y*, I6nn,N*,Y30n*

Y* - Use all reads in that cycle
I6nn - Drop last two reads in the cycle
N* - do not use any bases in that read
Y30n* - Use 30 bases drop the rest

That said unless you only need to specify this if you want to change the cycles numbers. If you want to process the run as set up then you can completely omit this parameter. bcl2fastq will find out the relevant information from RunInfo.xml file.

Answer 1

[swbarnes2]

Just what it says, it can't tolerate the "I*" in your --use-bases-mask part of the command.

I'd just drop it; the software should be able to figure everything out without it. I have the same version and I never include it, and the software figures it out.

Worst case, you run the software twice; once with the dual indexed samples, again with only the singles.

[drkennetz]

I generally run the software twice and specify a different report dir too if I am using a mix of single and dual-indexes. Something to keep in mind is if you are using SAV to view the metrics and you want to see the index distribution you should also specify a different interop path for one of the runs because the second run of the program will overwrite the first run if you do not specify.

[sourav.por]

Thank you for the suggestion. I actually add the --use-bases-mask for passing the difference between the 1st lane (6 base index) and the rest of all lane (8 base dual index). I should have tried without it. It's working now. Thanks.