Analysing BCR sequences

[ayeletperes]

Hey,

I've a three set of processed BCR sequences without primers:

1. Full VDJ region coverage without 5'UTR
2. Full VDJ region coverage with 5'UTR
3. Partial V coverage with full DJ coverage, without 5'UTR

I want to align the sequence to a custom germline library. Is this the correct way to align the sequences?

docker run --rm -e MI_LICENSE="${MI_LICENSE}" -v /work/:/work -v /work/mixcr/:/root/.mixcr/libraries/ ghcr.io/milaboratory/mixcr/mixcr:4.6.0 mixcr align -p generic-amplicon -s human -b database --rna --rigid-left-alignment-boundary --rigid-right-alignment-boundary J --keep-non-CDR3-alignments -OvParameters.geneFeatureToAlign="VRegion"

example sequences:

>Seq_1_set1
CAGGTGCAGCTGGCCCACTCTCGGACTCACATTAAGAACCCTGTTGCCTCAGGGAAGGTTTCCTCCAAGGCTACCGGTTACTGCTTCACTAGTTATGCTAGGAACTGGGTGCGAAAGGCCCGTGGACAAGGGGTTGTATGGATGGGATGGATCAACAACTACAAAGCGAAACCATCGTATGCACAGGGGTTCACCGGACTGAGCGTCTTCTCCTCGCACTCCTCTGTCAGCACGGCATATCTGCAGATCACCAACCTAAACGGTCAGGACACAGCCGTGTGTTACTGTGAGAGAGATTTTCCGCATCAGCACCGGGTCTACTACTACTACAATAAGCACACCTGGGGCCAAGGGACCATGGTCTGCATCTCCTCTT
>Seq_2_set2
CTGATTTCTGACAGCAATGTGATTCTCTGCGCATCACGACCAATCTGTCCCGGCTGAGGAGCGGCTGGTCGAGTCTGGAGGAGCCTGGGTCTAGCCGGGGGGGTCCCTGAGACTCTCCAGTAAAGGCGCTGGTTTCACCGTTAGTGGCGATGGGATGACCGGGGTCCGCCAGGCGCCACGGAAAGGGTTGGAATGGGTAGCCAACATAAAACAAGTCGGAAGTGTGAAATACTATGTGGACTCTGTGAAGGGACGATTGGACATCTACCAGAAAGAACCTCGCGGACTCACAGTAGCTCCAAATGAACAGCCTGAGAGCCGGGGAAACGGCTGTGATTAACTGTGCGAGCCTTCACCCTTCTGTCTATAGTGGCTCCGTGGGGCTACCTACGGCACTAGCGCCCGGGACCCTGGTCACTATCTCCTAAG
>Seq_3_set3
CTCTGGGTTTTCCCTCAGCACTAGTGGAATCCGTGTGAGCTGGNTCCGTCAGCCCCCAGGTCAGACCNGGGAGTGGCTTGCACGCANTGACTGGGATGATAATAAAGTCTACACCACATCTCTGAGNACCAGGCTTACCAACTCCGATGACACCTGCNAAAACCAGGTGCTCCTGAGAAGGAGCACCATGAACCCTGTGGACACAGCGACGTATNACTGTGCACGGACCCCCTAGACCCAAGCCATGTGCCGGTCACTGTAGTCGCCGCTTAGGTTCGACCCCTGGGGCCAGGGAACCCTGCTCACCGTCTCCTCGT

[mizraelson]

Hi,
Do you want to simply align the sequences and export the alignments for each read, or would you like to assemble the alignments into clonotypes? Also, have you already built the custom library for MiXCR?

[ayeletperes]

Hey,

I want to simply align the sequences and export the alignment. I don't need to assemble them into clones.

I created a custom library with the new AIRR-C reference set.

[ayeletperes]

This is the output

sequence_id	sequence	rev_comp	productive	v_call	d_call	j_call	c_call	sequence_alignment	germline_alignment	complete_vdj	junction	junction_aa	np1	np2	cdr1	cdr1_aa	cdr2	cdr2_aa	cdr3	cdr3_aa	fwr1	fwr1_aa	fwr2	fwr2_aa	fwr3	fwr3_aa	fwr4	fwr4_aa	v_score	v_cigar	d_score	d_cigar	j_score	j_cigar	c_score	c_cigar	junction_length	np1_length	np2_length	v_germline_start	v_sequence_start	v_germline_end	v_sequence_end	d_germline_start	d_sequence_start	d_germline_end	d_sequence_end	j_germline_start	j_sequence_start	j_germline_end	j_sequence_end	c_germline_start	c_sequence_start	c_germline_end	c_sequence_end	v_alignment_start	v_alignment_end	d_alignment_start	d_alignment_end	j_alignment_start	j_alignment_end	c_alignment_start	c_alignment_end
read.0	CAGGTCCAGCTGGTGCAGTCTGGGGCTGAGGAGAAGAAGCCTGGGTCCTCGGTGAAGGTCTCCTGCAAGGCTTCTGGAGGCACCTTCAGCAGCTATGCTATCAGCTGGGTGCGACAGGCCCCTGGACAAGGGCTTGAGTGGATGGGAAGGATCAACCCTATCTTTGGTACAGCAAACTACGCACAGAAGTTGCAGGGCAGAGTTACGATTACCGCGGACGAATCCACGAGCACAGTCTACATGGAGCTGAGCAGCCTGAGATCTCAGGACACGGCCGTGTATTACTGTGCTGCAGTCAGGCCCTACTGGAACTACTACTACGGTATGGATGTCTGGGGCCAATGGGCCAAGGTCACCGTCTCTTCAG	F	F	IGHVF6-G21*07	IGHD2-21*02	IGHJ6*02		CAGGTCCAGCTGGTGCAGTCTGGGGCTGAGGAGAAGAAGCCTGGGTCCTCGGTGAAGGTCTCCTGCAAGGCTTCTGGAGGCACCTTCAGCAGCTATGCTATCAGCTGGGTGCGACAGGCCCCTGGACAAGGGCTTGAGTGGATGGGAAGGATCAACCCTATCTTTGGTACAGCAAACTACGCACAGAAGTTGCAGGGCAGAGTTACGATTACCGCGGACGAATCCACGAGCACAGTCTACATGGAGCTGAGCAGCCTGAGATCTCAGGACACGGCCGTGTATTACTGTGCTGCAGTCAGGCCCTACTGGAACTACTACTACGGTATG	CAGGTCCAGCTGGTGCAGTCTGGGGCTGAGGTGAAGAAGCCTGGGTCCTCGGTGAAGGTCTCCTGCAAGGCTTCTGGAGGCACCTTCAGCAGCTATGCTATCAGCTGGGTGCGACAGGCCCCTGGACAAGGGCTTGAGTGGATGGGAAGGATCATCCCTATCTTTGGTACAGCAAACTACGCACAGAAGTTCCAGGGCAGAGTCACGATTACCGCGGACGAATCCACGAGCACAGCCTACATGGAGCTGAGCAGCCTGAGATCTGAGGACACGGCCGTGTATTACTGTGCNGCAGTCANNNNNNNNNNNNACTACTACTACGGTATG	F			T	GGCCCTACTGGA	GGAGGCACCTTCAGCAGCTATGCT	GGTFSSYA	ATCAACCCTATCTTTGGTACAGCA	INPIFGTA			CAGGTCCAGCTGGTGCAGTCTGGGGCTGAGGAGAAGAAGCCTGGGTCCTCGGTGAAGGTCTCCTGCAAGGCTTCT	QVQLVQSGAEEKKPGSSVKVSCKAS	ATCAGCTGGGTGCGACAGGCCCCTGGACAAGGGCTTGAGTGGATGGGAAGG	ISWVRQAPGQGLEWMGR	AACTACGCACAGAAGTTGCAGGGCAGAGTTACGATTACCGCGGACGAATCCACGAGCACAGTCTACATGGAGCTGAGCAGCCTGAGATCTCAGGACACGGCCGTGTATTAC	NYAQKLQGRVTITADESTSTVYMELSSLRSQDTAVYY			2726	31=1X122=1X36=1X11=1X31=1X28=1X25=	35	291S6N7=	170	310S26N17=				1	12	1	1	290	290	7	292	13	298	27	311	43	327					1	290	292	298	311	327

I've attached the report test.report.json

These were the commands:

docker run --rm -e MI_LICENSE="${MI_LICENSE}" -v /work/:/work -v /work/mixcr/:/root/.mixcr/libraries/ ghcr.io/milaboratory/mixcr/mixcr:4.7.0 mixcr align -p generic-amplicon --assemble-clonotypes-by CDR3 -s human -b database --rna --keep-non-CDR3-alignments --floating-left-alignment-boundary --rigid-right-alignment-boundary C --json-report /work/test.report.json --threads 20 --write-all -f /work/test.fasta /work/test.vdjca

docker run -e MI_LICENSE="${MI_LICENSE}" -v /work/:/work -v /work/mixcr/:/root/.mixcr/libraries/ ghcr.io/milaboratory/mixcr/mixcr:4.7.0 mixcr exportAirr -f /work/test.vdjca /work/test.tsv

[ayeletperes]

The --assemble-clonotypes-by CDR3 flag does not affect the align command output. The change in the number of hits is expected as we now export a single -vHit instead of all -vHits by default. That is to make the exportAirr output more compatible with other softwares.

Is the alignment command perform any correction to the sequences, though they are without UMIs?

[mizraelson]

Hi,

Sorry for the delay. We have looked into the issue with this sequence and found a bug that results in labeling this clone as non-functional. We will fix this in the upcoming release.

Please note that the align command does not perform any corrections.

[ayeletperes]

Hey,

Thank you for the update. I think I might have found another bug. I've used the same command to run the sequence below, MiXCR has tagged the sequence as productive though in the translation of the cdr2 region (cdr2_aa) there is a stop codon.

sequence_id	sequence	rev_comp	productive	v_call	d_call	j_call	c_call	sequence_alignment	germline_alignment	complete_vdj	junction	junction_aa	np1	np2	cdr1	cdr1_aa	cdr2	cdr2_aa	cdr3	cdr3_aa	fwr1	fwr1_aa	fwr2	fwr2_aa	fwr3	fwr3_aa	fwr4	fwr4_aa	v_score	v_cigar	d_score	d_cigar	j_score	j_cigar	c_score	c_cigar	junction_length	np1_length	np2_length	v_germline_start	v_sequence_start	v_germline_end	v_sequence_end	d_germline_start	d_sequence_start	d_germline_end	d_sequence_end	j_germline_start	j_sequence_start	j_germline_end	j_sequence_end	c_germline_start	c_sequence_start	c_germline_end	c_sequence_end	v_alignment_start	v_alignment_end	d_alignment_start	d_alignment_end	j_alignment_start	j_alignment_end	c_alignment_start	c_alignment_end
read.3	GAGAAGCAGCTTGTGGAGACTGGGCGAGGCTTGGCACAGCCAGCGCGCTCCCTGAGACTCTCCTGTTGACCTTCTGGATTCATCTTGGGTGATTATGCTATGAGCTGGTTACGCCAGGCTTGAGGGAAGGGGCTGGAGTGTGTAAATTTC	F	T	IGHVF8-G30*02	IGHD1-1*01	IGHJ6*03	NaN	GCAGCTTGTGGAGACTGGGCGAGGCTTGGCACAGCCAGCGCGCTCCCTGAGACTCTCCTGTTGACCTTCTGGATTCATCTTGGGTGATTATGCTATGAGCTGGTTACGCCAGGCTTGAGGGAAGGGGCTGGAGTGTGTAAATTTC	GCAGCTGGTGGAGTCTGGGGGAGGCTTGGTACAGCCAGGGCGGTCCCTGAGACTCTCCTGTACAGCTTCTGGATTCACCTTTGGTGATTATGCTATGAGCTGGTTCCGCCAGGCTC	F	GGTACTAGAGATATCTTCGGAACGACCACTTACATGGACGTCTGG	GTRDIFGTTTYMDVW	TATCTTC	CACT	GGATTCATCTTGGGTGATTATGCT	GFILGDYA	GTTTGAATCAAAGCTTAAGGTGGAACAGCA	V*IKA*GGTA	ACTAGAGATATCTTCGGAACGACCACTTACATGGACGTC	TRDIFGTTTYMDV	NaN	NaN	ATGAGCTGGTTACGCCAGGCTTGAGGGAAGGGGCTGGAGTGTGTAAATTTC	MSWLRQA*GKGLECVNF	GAATACGCCTCCTCTGTGTCAGGCAAATACACCATCTCAAGAGATCAAACCCAAAGCATGGCCCATCTGCAAATGAACAGCCTGAAAACCGAGGACACGGCCGTAAATTAC	EYASSVSGKYTISRDQTQSMAHLQMNSLKTEDTAVNY	GGTAAACTGATCACACTCACCTTCCCCTCAG	GKLITLTFPS_	1860.0	5S5N6=1X6=1X5=1X9=1X8=1X3=1X18=2X1=1X12=1X3=1X23=1X9=2X18=1X3=2X4=1X2=1X3=1X9=1X5=1X3=1X11=1X1=1X6=2X5=1X2=1X16=1X1=	40.0	309S26N8=	228.0	321S37N17=1X3=2X2=1X3=2X5=1X2=1X6=	NaN	NaN	45.0	7.0	4.0	6.0	6.0	307.0	307.0	27.0	310.0	34.0	317.0	38.0	322.0	83.0	367.0	NaN	NaN	NaN	NaN	1.0	302.0	305.0	312.0	317.0	362.0	NaN	NaN

[mizraelson]

This is a slightly different case. By default, the authentic AIRR format returns a productive/non-productive value based on the CDR3 sequence. To be consistent with the format in the exportAirr command, we used the same logic, so the sequence is marked as functional or non-functional based on the CDR3. In the regular MiXCR output (not AIRR), you can add a similar column by specifying the feature to use. For example:

-isProductive VDJRegion

This will add a column with a boolean value based on the entire VDJRegion sequence.