Amino acid mutations/alignments with reference genes

milaboratory / mixcr

MiXCR is an ultimate software platform for analysis of Next-Generation Sequencing (NGS) data for immune profiling.

https://mixcr.com

Other

323 stars 78 forks source link

Amino acid mutations/alignments with reference genes #4

Closed dbolotin closed 7 years ago

dbolotin commented 9 years ago

From this letter:

Oh well. One more question about the mutations: these are interpretable as SHM, right? Assuming there is not sequencing/pcr error, so NGS being completely error-free, then these mutations would be SHM, and, not as it is now the case a mixture of SHM and NGS-related errors, right? And a suggestion: it would be nice to have them also on the amino acid levels (similar to IMGT).

mikessh commented 9 years ago

Complex issue. For example how to report two consequent substitutions, which are only non-synonymous when present together. Either report them as synonymous or allow to aggregate mutations (milib)

dbolotin commented 9 years ago

I'm thinking about simplest algorithm:

Translate mutated sequence
Enumerate substitutions between mutated and germline sequences


ATTAAGACACAGATA
||||  |||| ||||
ATTATAACACCGATA

->

ATTAAGACACAGATA
 I  K  T  Q  I
 |     |     |
 I  I  T  P  I
ATTATAACACCGATA

->

SK1I, SQ3P

Exceptional cases:

If sequence contain long (~ >15 nt) region with shifted frame, don't output mutations, just print something like (frameshift)
But if region is not too long (like the following), still print mutations for it.


ATTAAGACACA-GATA
ATTA-GACACATGATA

->

ATTAAGACACAGATA
 I  K  T  Q  I
 |           |
 I  R  H  M  I
ATTAGACACATGATA

->

SK1R, ST2H, SQ3M

For deletions of 3,6,9... nucleotides there is also possible to output amino acid deletions... But it is not so simple in general case

dbolotin commented 8 years ago

Basic feature for this is already implemented in com.milaboratory.core.mutations.MutationsUtil.nt2aa()...

swuecho commented 8 years ago

"it is now the case a mixture of SHM and NGS-related"

NGS error should be negligible compared with shm, since 'the mutation occur at a rate of about one muation per V-region sequence per cell division" ?

I am trying to calculated SHM using mixcr.

I use the command

mixcr exportClones -vHit -jHit -dHit -count -aaFeature CDR3 -vAlignment -jAlignment

to export the clone with v j alignment, and calculate the mutation based on alignment. but the shm result is very different from what I get using igblast. (we are switching from igblast to mixcr)

Am I in the right direction? Thanks.

of course, if the shm can be exported directly, it will be great.