The loss of methylation syntax returns unexpected error.

[ifokkema]

According to the nomenclature pages, the pipe is used for:

(...) indicate that not a direct change of the sequence is described but a modification (a change of state, e.g. methylation). (see Example methylation)

This example shows:

“gom” indicates a gain of methylation; g.12345678_12345901|gom “lom” indicates a loss of methylation; g.12345678_12345901|lom “met” indicates a methylation; g.12345678_12345901|met=

Before I'm pretty sure I have seen syntax errors with these variants, but now it seems the pipe is translated into a separator of variants. Submitting NC_000011.9:g.2018812_2024740|lom returns some (for me) unexpected results:

{
  "NC_000011.9:g.2018812_2024740": {
    "NC_000011.9:g.2018812_2024740": {
      "genomic_variant_error": "NC_000011.9:g.2018812_2024740: char 30: end of input"
    },
    "errors": [],
    "flag": "genomic_variant_warning"
  },
  "lom": {
    "errors": [],
    "flag": "submission_warning",
    "lom": {
      "genomic_variant_error": "Variant description lom is not in a supported format"
    }
  }
}

The variant should not be split in two; if the pipe is not supported, it should throw a syntax warning.

[Peter-J-Freeman]

We do use Pipe internally as a separator. Forgot about its use here. Seriously, can this nomenclature not be simplifed. It's getting crazy!!

Haha

[ifokkema]

I wouldn't mind a simplification, but it looks like new situations seem to cause new characters being included in the description :roll_eyes:

[Peter-J-Freeman]

It's something that needs to be addressed. It is making computation more and more difficult!

[ifokkema]

This still applies.

[Peter-J-Freeman]

Yes it does. It's currently pretty low priority though. I'm sick of the HGVS using every available ascii character needlessly. There is no need for the | in the gom lom desccriptions

[ifokkema]

OK, we can then fix this by writing a wrapper function that sends the variant as an = and then converts the reply back to |gom when needed.

[Peter-J-Freeman]

Think we could do the same to be honest. Needs plugging in though. Can you please raise it in the LOVD alignment issue mate? Want to capture all these issues in one place. This is an easy one for the students to complete

[Peter-J-Freeman]

Actually, I will add it.

[ifokkema]

Alright! Great! I just expected it to be an issue as it conflicts with your batch feature. So it seems either the batch feature works or these variants would be supported. Hence my thought was to just "trick" VV with a different variant and keep it LOVD-side.

[Peter-J-Freeman]

Nope, that's the point of this issue and exercise. I have students writing code to start handling these variants. Better you help them rather than make needless fudges. Keep your eye on issues