Genome Nexus: 41 variants that have incorrectly annotated ref/alt.

[thomasyu888]

Uploaded as file for ease. ref_alt_diff.txt

[inodb]

@thomasyu888 could you elaborate what these columns represent exactly:

CURRENT.START   CURRENT.REF     CURRENT.ALT     EXPECTED.START  EXPECTED.REF    EXPECTED.ALT

I assume that we need at least these three sets of columns to solve this issue (or something similar):

INPUT
CURRENT_OUTPUT
EXPECTED_OUTPUT

But I only see EXPECTED and CURRENT

In addition these miss end position. In my understanding the annotation-tools converts VCF to MAF and thereby computes the End_Position. It would be good to add that as well. Ideally to make the debugging easier, we would have all this:

INPUT_ANNOTATION_TOOLS
OUTPUT_ANNOTATION_TOOLS
OUTPUT_GENOME_NEXUS_ANNOTATON_PIPELINE
EXPECTED_OUTPUT_GENOME_NEXUS_ANNOTATION_PIPELINE

Could you help provide these?

[thomasyu888]

Ah, @inodb this is only the current output (CURRENT...) and expected output (EXPECTED...)

I didn't pull down the input. One thing to note is that if these are from maf files, they may not have end_positions. I would have to double check.

[thomasyu888]

Sorry for the delay, Ill slowly add resources here. There is one variant that is a VCF. I won't upload the ref_alt_diff.txt file again, but here is.

center input: input.vcf.txt
Annotation Pipeline initial output: input.vcf.temp.txt
Genome Nexus output: input.vcf.temp.annotated.txt
Annotation Pipeline final: processed.txt

[thomasyu888]

More resources - I split these up because some sites have different input files as well.

center input: input_1.txt
Annotation Pipeline initial output: input.txt.temp.txt
Genome Nexus output: input.txt.temp.annotated.txt
Annotation Pipeline final: processed.txt

[thomasyu888]

More:

center input: input_2.txt
Annotation Pipeline initial output: input.txt.temp.txt
Genome Nexus output: input.txt.temp.annotated.txt
Annotation Pipeline final: processed.txt

[thomasyu888]

Final:

center input: input.txt
Annotation Pipeline initial output: input.txt.temp.txt
Genome Nexus output: input.txt.temp.annotated.txt
Annotation Pipeline final: processed.txt

[ao508]

@thomasyu888 I've started looking into this and wanted to follow up on an inconsistency I found. Idk if there are others, just that the first one I started looking at the REF/ALT are not what's reported in the diff file you shared above.

These are the POS, REF, ALT in the VCF from the input file in this comment

current pos	current ref	current alt
7578397	GAA	TCC

But the diff file shows

current pos	current ref	current alt
7578397	TGG	TCC

[ao508]

I found similar inconsistencies in the other input example files as well. I'm not really sure how to approach this either because of the differing inputs

from the diff file shared:

pos	ref	alt
7578535	TTGTTGAGGGCAGGGGAGTAC	TGAGGGCAGGGGAGTA

from the input file (input2)

pos	ref	alt
7578533	ATCTTGTTGAGGGCAGGGGAGTA	ATTGAGGGCAGGGGAGTA

Just eye balling what this might resolve to.. which still doesn't match what the ref_alt_diff.txt has for this variant.

my assumption:

pos	ref	alt
7578535	CTTGTTGAGGGCAGGGGAGTA	TGAGGGCAGGGGAGTA

the expected values reported in the diff file shared

pos	ref	alt
7578536	TGTTGAGGGCAGGGGAGTAC	GAGGGCAGGGGAGTA

[thomasyu888]

Sorry about the confusion @ao508 . The diff file doesnt show what's in the input.txt. The diff file is supposed to match what's in processed.txt. what the collaborator is suggesting is that those are incorrect and the corrected ref/alt in the other columns of the diff file.

[ao508]

I think I've resolved an issue that affected at least some of these variants #39