Amino acid mutations over time: invalid mutation code

[chaoran-chen]

On the influenza H1N1pdm dashboard, the amino acid mutations over time component is failing:

The component is asking for NP:X114D in its request to the aminoAcidMutationsOverTime-endpoint but that doesn't make sense as X stands for unknown and the reference genome shouldn't have unknowns.

[fhennig]

I had a look at this, and I found that in the request that fetches all mutations, the mutations are already there.

request:

curl 'https://api.loculus.genspectrum.org/h1n1pdm/sample/aminoAcidMutations' \
  -X POST \
  -H 'Content-Type: application/json' \
  --data-raw '{"sampleCollectionDateFrom":"2025-01-13","versionStatus":"LATEST_VERSION","isRevocation":"false","sampleCollectionDateTo":"2026-01-04","minProportion":0.001}'

part of the response:

...
    {
      "mutation": "NP:X114D",
      "count": 21,
      "coverage": 6091,
      "proportion": 0.003447709735675587,
      "sequenceName": "NP",
      "mutationFrom": "X",
      "mutationTo": "D",
      "position": 114
    },
    {
      "mutation": "NP:X114E",
      "count": 6070,
      "coverage": 6091,
      "proportion": 0.9965522902643245,
      "sequenceName": "NP",
      "mutationFrom": "X",
      "mutationTo": "E",
      "position": 114
    },
...

When I fetch the reference genome, it actually does contain an X!

MASQGTKRSYEQMETGGERQDATEIRASVGRMIGGIGRFYIQMCTELKLSDYDGRLIQNSITIERMVLSAFDERRNKYLEEHPSAGKDPKKTGGPIYRRVDGKWMRELILYDKXEIRRVWRLANNGEDATAGLTHIMIWHSNLNDATYQRTRALVRTGMDPRMCSLMQGSTLPRRSGAAGAAVKGVGTIAMELIRMIKRGINDRNFWRGENGRRTRVAYERMCNILKGKFQTAAQRAMMDQVRESRNPGNAEIEDLIFLARSALILRGSVAHKSCLPACVYGLAVASGHDFEREGYSLVGIDPFKLLQNSQVVSLMRPNENPAHKSQLVWMACHSAAFEDLRVSSFIRGKKVIPRGKLSTRGVQIASNENVETMDSNTLELRSRYWAIRTRSGGNTNQQKASAGQISVQPTFSVQRNLPFERATVMAAFSGNNEGRTSDMRTEVIRMMESAKPEDLSFQGRGVFELSDEKATNPIVPSFDMSNEGSYFFGDNAEEYDS*

( https://api.loculus.genspectrum.org/h1n1pdm/sample/referenceGenome?downloadAsFile=false )

I'm not quite sure where this is coming from.

[fhennig]

Alright! So it turns out that this X is actually in the reference sequence, and it's correct: https://www.ncbi.nlm.nih.gov/nuccore/NC_026436.1

This means, the way the issue is written at the moment ("invalid mutation code") isn't right.

I think instead, we need to somehow make the component more resilient, and allow X in the reference.