bcl2fastq discard complete read

Question

[DCZ]

Hi,

I'm trying to demultiplex a run with Read settings: 28-8-91.

However, I don't need read 1 for this. Therefore, I set --use-bases-mask N*,I*,Y*.

Nevertheless, this consistently fails with the following output:

2020-03-25 15:43:25 [7f093bf05700] WARNING: Mismatching cluster count in positions file: Cycle #1, Tile#1101: bytes_read=4091904 bytes_expected=2988132
2020-03-25 15:43:25 [7f093bf05700] WARNING: Mismatching cluster count in filter file: Cycle #1, Tile#1101: bytes_read=4091904 bytes_expected=2988132
2020-03-25 15:43:26 [7f093df09700] WARNING: Mismatching cluster count in positions file: Cycle #1, Tile#1102: bytes_read=2988132 bytes_expected=2997896
2020-03-25 15:43:26 [7f093df09700] WARNING: Mismatching cluster count in filter file: Cycle #1, Tile#1102: bytes_read=4091904 bytes_expected=2997896
2020-03-25 15:43:26 [7f093bf05700] WARNING: Mismatching cluster count in positions file: Cycle #1, Tile#1103: bytes_read=2997896 bytes_expected=2996204
2020-03-25 15:43:26 [7f093bf05700] WARNING: Mismatching cluster count in filter file: Cycle #1, Tile#1103: bytes_read=4091904 bytes_expected=2996204
2020-03-25 15:43:27 [7f093a702700] ERROR: Thread: 8 caught an exception first: Dynamic exception type: std::out_of_range
std::exception::what: vector::_M_range_check: __n (which is 1) >= this->size() (which is 1) terminate called after throwing an instance of 'boost::exception_detail::clone_impl<boost::exception_detail::current_exception_std_exception_wrapper<std::out_of_range>
  what():  vector::_M_range_check: __n (which is 1) >= this->size() (which is 1) Aborted (core dumped)

It seems like it's impossible to discard an entire read. If i change the basemask to YN*,I*,Y* everything works fine. I can easily solve it by just discarding R1 afterwards, but i'm just wondering why this behaviour exists.

Kind regards

[GenoMax]

Try --use-bases-mask n*,I*,Y*.

This appears to be a 10x dataset. You should ideally use cellranger mkfastq option for the Cellranger software to demultiplex this data. You can download from 10x Genomics here. Free registration required.

[DCZ]

Hi Genomax. That's what i did, the formatting got lost. i used --use-bases-mask N*,I*,Y*. Sorry for that. I tried it both with capital N & normal n. Indeed it's 10x. But also other constructs in the pool. The 10x demultiplexing with cellranger went fine.

[GenoMax]

I think you need to use a lower case n*. Can you check that?

[DCZ]

I had tried that before. Tried it again to no avail:

2020-03-25 16:48:36 [7f311c418700] WARNING: Mismatching cluster count in positions file: Cycle #1, Tile#1101: bytes_read=4091904 bytes_expected=2988132
2020-03-25 16:48:36 [7f311c418700] WARNING: Mismatching cluster count in filter file: Cycle #1, Tile#1101: bytes_read=4091904 bytes_expected=2988132
2020-03-25 16:48:37 [7f311e41c700] WARNING: Mismatching cluster count in positions file: Cycle #1, Tile#1102: bytes_read=2988132 bytes_expected=2997896
2020-03-25 16:48:37 [7f311e41c700] WARNING: Mismatching cluster count in filter file: Cycle #1, Tile#1102: bytes_read=4091904 bytes_expected=2997896
2020-03-25 16:48:37 [7f311e41c700] WARNING: Mismatching cluster count in positions file: Cycle #1, Tile#1103: bytes_read=2997896 bytes_expected=2996204
2020-03-25 16:48:37 [7f311e41c700] WARNING: Mismatching cluster count in filter file: Cycle #1, Tile#1103: bytes_read=4091904 bytes_expected=2996204
2020-03-25 16:48:37 [7f311f41e700] ERROR: Thread: 3 caught an exception first: Dynamic exception type: std::out_of_range
std::exception::what: vector::_M_range_check: __n (which is 1) >= this->size() (which is 1)

terminate called after throwing an instance of 'boost::exception_detail::clone_impl<boost::exception_detail::current_exception_std_exception_wrapper<std::out_of_range> >'
  what():  vector::_M_range_check: __n (which is 1) >= this->size() (which is 1)
Aborted (core dumped)

Answer 1

[GenoMax]

I just tested this. Looks like you have to process the first read in order for bcl2fastq to work. So no way around that.

[DCZ]

Ok. Strange though! Thanks for your time!