calling a deletion in IGK-J with igbalst 1.12.0

[HLAmolec]

Hello,

igblastn 1.12.0 is not calling a deletion in IGK-J with the following reference. blastn has no problem calling it at the default settings. I've tried to adjust the parameters to match blastn but the deletion was not called. Igblast returns a blank for J-segment. Do you have any suggestions how to make this work with igblast?

Thank you for your help.

here is the queried sequence:

>test_sequence
AGAGCCGCGGTCTTTCTCGATTGAGTGGCTTTGGTGGCCATGCCACCGCGCTCTTGGGGCAGCCGCCTTGCCGCTAGTGGCCGTGGCCACCCTGTGTCTGCCCGATTGATGCTGCCGTAGCCAGCTTTCCTGAGCCCTAGTGGCAGCCCAGGGCGACTCCTCATGAGTCTGCAGCTGCATTTTTGCCATATCCACTATTTGGAGTCTGACCTCCCTAGGAAGCCTCCCTGCTCCC

Here is the reference sequence:

>IGKDEL
GGTACCTGAAGAATTTTCAGGATCAACATTTGTAAGCATTATCAGCCCAGGATAATGCTTATTTACAGCAGTTCAACTTTTCTCACCTGCAGTAGTTTACTTTGATGAAATCCTGCTTATTTTTCACAATTAAAATTGCGTCCATGGCCCTTGTTACCCTTAAGAATTATATTTTGCTGACACAATCTCTTTATCTACATTTCACCTAAGTAAGCTTTTTCTTTACATGATTTTAAGTCATTTTCCAAAACAGAAAAGCATAATTATACAACCTTCTATGTGTATTGGAGGAAAAAAAGGGTCACACAAATAATAGTACACTCTCTCAGTTAAGGAGTAGGCAGGAGGAGGGGGTTGAGAAAGACAAACCAGAAAGAATTACATGAAATTTTCCTTCCTCCTCACTGAGCCTCCCTTGAATTGTCCAATAATATCCCCCCAACACACACACATAGACACAGAGCAGACTGTGAGGGACACGCAGCCTGGGTGGGACTCCAGGAGCCAGCTCTTACCCTAGAGTTTCTGCACGGGCAGCAGGTTGGCAGCGCACACTGTGGGAGCCCTAGTGGCAGCCCAGGGCGACTCCTCATGAGTCTGCAGCTGCATTTTTGCCATATCCACTATTTGGAGTCTGACCTCCCTAGGAAGCCTCCCTGCTCCCTAGGACAACCTGCTCTGACCTCTGAGGACCTGTCTGTAAACGTCCAGAGAAAAGCATGTGCCTGAAGGGTCTATGAAGGGGCTTGAGGCAAGTAGGGAGCCCAGCCCAGCTAACATTTGCAGCCATGGGATGGCTTTGTGTACCTAGAAAAGCAAAGATGAGGACTAGGCGAGCACAGGCCCCAAACGTTCACGATACACATCTCACTGCAAAGAAATGTCCTCTGGCCACTTTATAATGCAGCTCTAACTATAACTGGTGCTTTGCTGGTTTGTGGCCTGGGCTGGTCTCCCAGAGTCAGTGGCTTTGGGTGAGATGGCTCCAGGAGACAGCAGAAACTCTCATATATGAAGCCTTGCTTGCAGCGATTTGAGGCTTACTAGGGAAAAGCCATGATGGGTTTTATAGAGCATTAACTGTGATACGCCATCTAGGGCTGAGAGCTC

[mstubb]

Hi,

I'm a bit confused about what you're trying to do here. Please could you elaborate a bit?

Is this a deletion that you're observing in some of your cells? Are you trying to use a custom reference? Is this mouse or human?

Thanks,

Mike

On 11 Jun 2019, at 15:26, Abracodabra notifications@github.com wrote:

Hello,

igblastn 1.12.0 is not calling a deletion in IGK-J with the following reference. blastn has no problem calling it at the default settings. I've tried to adjust the parameters to match blastn but the deletion was not called. Igblast returns a blank for J-segment. Do you have any suggestions how to make this work with igblast?

Thank you for your help.

here is the queried sequence:

test_sequence AGAGCCGCGGTCTTTCTCGATTGAGTGGCTTTGGTGGCCATGCCACCGCGCTCTTGGGGCAGCCGCCTTGCCGCTAGTGGCCGTGGCCACCCTGTGTCTGCCCGATTGATGCTGCCGTAGCCAGCTTTCCTGAGCCCTAGTGGCAGCCCAGGGCGACTCCTCATGAGTCTGCAGCTGCATTTTTGCCATATCCACTATTTGGAGTCTGACCTCCCTAGGAAGCCTCCCTGCTCCC Here is the reference sequence:

IGKDEL GGTACCTGAAGAATTTTCAGGATCAACATTTGTAAGCATTATCAGCCCAGGATAATGCTTATTTACAGCAGTTCAACTTTTCTCACCTGCAGTAGTTTACTTTGATGAAATCCTGCTTATTTTTCACAATTAAAATTGCGTCCATGGCCCTTGTTACCCTTAAGAATTATATTTTGCTGACACAATCTCTTTATCTACATTTCACCTAAGTAAGCTTTTTCTTTACATGATTTTAAGTCATTTTCCAAAACAGAAAAGCATAATTATACAACCTTCTATGTGTATTGGAGGAAAAAAAGGGTCACACAAATAATAGTACACTCTCTCAGTTAAGGAGTAGGCAGGAGGAGGGGGTTGAGAAAGACAAACCAGAAAGAATTACATGAAATTTTCCTTCCTCCTCACTGAGCCTCCCTTGAATTGTCCAATAATATCCCCCCAACACACACACATAGACACAGAGCAGACTGTGAGGGACACGCAGCCTGGGTGGGACTCCAGGAGCCAGCTCTTACCCTAGAGTTTCTGCACGGGCAGCAGGTTGGCAGCGCACACTGTGGGAGCCCTAGTGGCAGCCCAGGGCGACTCCTCATGAGTCTGCAGCTGCATTTTTGCCATATCCACTATTTGGAGTCTGACCTCCCTAGGAAGCCTCCCTGCTCCCTAGGACAACCTGCTCTGACCTCTGAGGACCTGTCTGTAAACGTCCAGAGAAAAGCATGTGCCTGAAGGGTCTATGAAGGGGCTTGAGGCAAGTAGGGAGCCCAGCCCAGCTAACATTTGCAGCCATGGGATGGCTTTGTGTACCTAGAAAAGCAAAGATGAGGACTAGGCGAGCACAGGCCCCAAACGTTCACGATACACATCTCACTGCAAAGAAATGTCCTCTGGCCACTTTATAATGCAGCTCTAACTATAACTGGTGCTTTGCTGGTTTGTGGCCTGGGCTGGTCTCCCAGAGTCAGTGGCTTTGGGTGAGATGGCTCCAGGAGACAGCAGAAACTCTCATATATGAAGCCTTGCTTGCAGCGATTTGAGGCTTACTAGGGAAAAGCCATGATGGGTTTTATAGAGCATTAACTGTGATACGCCATCTAGGGCTGAGAGCTC

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/Teichlab/tracer/issues/92?email_source=notifications&email_token=AAIXBBRCQS5TOXBJSC7Y24DPZ6Y2XA5CNFSM4HW65VLKYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4GY2RGDQ, or mute the thread https://github.com/notifications/unsubscribe-auth/AAIXBBSNI45XGQWFLTKZGHTPZ6Y2XANCNFSM4HW65VLA.

[HLAmolec]

Sorry for the confusion. Correct, This is a deletion that is observed in some of the cells. It is a custom made human reference, which works great with blastn, however, igblast doesn't call it.

Thanks

[mstubb]

Ah ok. Am I correct that the test_sequence you sent below is the J gene only? IgBLAST expects to first find a V gene; if it can't do that then it won't work properly?

Do you have a rearranged IgK sequence containing a V gene upstream of the J with the deletion?

Best,

Mike

On 11 Jun 2019, at 15:56, Abracodabra notifications@github.com wrote:

Sorry for the confusion. Correct, This is a deletion that is observed in some of the cells. It is a custom made human reference, which works great with blastn, however, igblast doesn't call it.

Thanks

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/Teichlab/tracer/issues/92?email_source=notifications&email_token=AAIXBBXAE6JWMMDF5TZVLC3PZ64JVA5CNFSM4HW65VLKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODXNNCYY#issuecomment-500879715, or mute the thread https://github.com/notifications/unsubscribe-auth/AAIXBBXOJQZ5HXDJ5OUREWDPZ64JVANCNFSM4HW65VLA.

[HLAmolec]

The V segment for that sequence is intronic rearrangement. Also, a custom made reference that works well with igblast.

>IGKINTR
CGTGGCACCGCGAGCTGTAGACAGAGCCGCGGTCTTTCTCGATTGAGTGGCTTTGGTGGCCATGCCACCGCG
CTCTTGGGGCAGCCGCCTTGCCGCTAGTGGCCGTGGCCACCCTGTGTCTG
CCCGATTGATGCTGCCGTAGCCAGCTTTCCTG

[HLAmolec]

So, igblast returns IGKINTR for V but nothing for J.

[mstubb]

OK. So if you have sequences that contain IGKINTR–IGKDEL rearrangements and you're confident that your reference is correctly set up then I'm not sure why IgBLAST is failing. Maybe this is a question for the IGBLAST developers.

Good luck!

Mike

On 11 Jun 2019, at 16:06, Abracodabra notifications@github.com wrote:

So, igblast returns IGKINTR for V but nothing for J.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/Teichlab/tracer/issues/92?email_source=notifications&email_token=AAIXBBRMNK7GX73HRV2F3NTPZ65N5A5CNFSM4HW65VLKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODXNOE7Y#issuecomment-500884095, or mute the thread https://github.com/notifications/unsubscribe-auth/AAIXBBXO32LVL5G4HLV3RMDPZ65N5ANCNFSM4HW65VLA.

[HLAmolec]

What is the best way to contact the IgBlast developers? Thanks

[mstubb]

This page (https://ncbi.github.io/igblast/ https://ncbi.github.io/igblast/) suggests blast-help@ncbi.nlm.nih.gov

On 11 Jun 2019, at 16:08, Abracodabra notifications@github.com wrote:

What is the best way to contact the IgBlast developers? Thanks

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/Teichlab/tracer/issues/92?email_source=notifications&email_token=AAIXBBXXCM45BKJDVFO7EYDPZ65YVA5CNFSM4HW65VLKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODXNOONA#issuecomment-500885300, or mute the thread https://github.com/notifications/unsubscribe-auth/AAIXBBTBBCPWYFGXSENHXMDPZ65YVANCNFSM4HW65VLA.