transanno liftvcf "Error: length of chromosome [chr] is not equal to length in chain file. Are you using correct reference?"

Question

[ezz3]

Hello Everyone,

I am trying to do a liftover with mitochondrial sequences using transanno and am running into issues.

I created my own chain file using .fa reference sequences and following this tutorial and it works and gives expected coordinates when I use picard LiftOverVcf.

picard LiftoverVcf -I $VCFIN -O $VCFOUT -C $CHAIN -R $REF --REJECT $REJECTS

Where:

$VCFIN is my vcf called from reads aligned to old build coordinates

$VCFOUT is the lifted over output vcf

$CHAIN is the chain file

$REF is the new build .fa file

$REJECTS is the variants that fail to liftover

My chain file is:

##matrix=axtChain 16 91,-114,-31,-123,-114,100,-125,-31,-31,-125,100,-114,-123,-31,-114,91

##gapPenalties=axtChain O=400 E=30

chain 1556846 chrM 16567 + 0 16567 chrM 16569 + 0 16569 1

309 1 2

212 2 0

15734 2 1

4 0 1

213 2 1

90

Where the first 16567 bp sequence is the target/reference (source) and the second 16569 is the query (destination).

When I try to run transanno:

transanno liftvcf -m -c $CHAIN \
-o $VCFOUT \
-q $REF\
-r $SOURCE \
--vcf $VCFIN \
--fail $REJECTS

Where $SOURCE is the .fa of the source reference genome.

I get this error: "Error: length of chromosome chrM is not equal to length in chain file. Are you using correct reference?"

Does anyone have any suggestions?

I tried changing the names of the target (source) sequence and the query (destination) sequence and reversing reference and query since the naming is somewhat confusing but neither resulted in a successful liftover. I also tried going into the command source code but have not been successful.

Thanks!

[Pierre Lindenbaum]

"Are you using correct reference?"

may be your reference is not using the same chrM. see diferent contigs chrM

[ezz3]

Thanks. I was wondering if that is the case but was having some issues understanding the transanno code which is written in rust and I don't see any arguments to easily change this because I would like to use the two .fa sequences I input into the liftvcf command for my analysis. I am testing out other liftover programs but was hoping to get transanno liftvcf working.