Jannovar does not annotate deletions for stop_gained (only frameshift)

[janeliutw]

Describe the bug

Bug 1: A nonsense indel in Jannovar is annotated as frameshift_variant. VEP also calls it frameshift_variant. However, pdots are not the same.

For example,
annotate-var/refseq/hg19/chr1/197071665/AGTTT/A

[
   {
      "transcriptId":"NM_018136.5",
      "variantEffects":[
         "frameshift_truncation"
      ],
      "isCoding":true,
      "hgvsProtein":"p.(K2238fs)",
      "hgvsNucleotides":"c.6712_6715del"
   },
   {
      "transcriptId":"NM_001206846.2",
      "variantEffects":[
         "coding_transcript_intron_variant"
      ],
      "isCoding":true,
      "hgvsProtein":"p.(=)",
      "hgvsNucleotides":"c.4066-6375_4066-6372del"
   }
]

VEP results

VEP indicates p. should be p.Lys2238Ter while Jannovar returns p.(K2238fs)

Bug 2: Nonsense indel gets disruptive_inframe_deletion from Jannovar but stop_gained from VEP.

For example,
annotate-var/refseq/hg19/chr8/38277141/CTTG/C

[
   {
      "transcriptId":"NM_001174063.2",
      "variantEffects":[
         "disruptive_inframe_deletion"
      ],
      "isCoding":true,
      "hgvsProtein":"p.(Y397_K398delins*)",
      "hgvsNucleotides":"c.1191_1193del"
   },
   {
      "transcriptId":"NM_001174064.2",
      "variantEffects":[
         "disruptive_inframe_deletion"
      ],
      "isCoding":true,
      "hgvsProtein":"p.(Y389_K390delins*)",
      "hgvsNucleotides":"c.1167_1169del"
   },
   {
      "transcriptId":"NM_001174065.2",
      "variantEffects":[
         "disruptive_inframe_deletion"
      ],
      "isCoding":true,
      "hgvsProtein":"p.(Y395_K396delins*)",
      "hgvsNucleotides":"c.1185_1187del"
   },
   {
      "transcriptId":"NM_001174066.2",
      "variantEffects":[
         "disruptive_inframe_deletion"
      ],
      "isCoding":true,
      "hgvsProtein":"p.(Y308_K309delins*)",
      "hgvsNucleotides":"c.924_926del"
   },
   {
      "transcriptId":"NM_001174067.2",
      "variantEffects":[
         "disruptive_inframe_deletion"
      ],
      "isCoding":true,
      "hgvsProtein":"p.(Y428_K429delins*)",
      "hgvsNucleotides":"c.1284_1286del"
   },
   {
      "transcriptId":"NM_001354367.2",
      "variantEffects":[
         "disruptive_inframe_deletion"
      ],
      "isCoding":true,
      "hgvsProtein":"p.(Y395_K396delins*)",
      "hgvsNucleotides":"c.1185_1187del"
   },
   {
      "transcriptId":"NM_001354368.2",
      "variantEffects":[
         "disruptive_inframe_deletion"
      ],
      "isCoding":true,
      "hgvsProtein":"p.(Y306_K307delins*)",
      "hgvsNucleotides":"c.918_920del"
   },
   {
      "transcriptId":"NM_001354369.2",
      "variantEffects":[
         "disruptive_inframe_deletion"
      ],
      "isCoding":true,
      "hgvsProtein":"p.(Y395_K396delins*)",
      "hgvsNucleotides":"c.1185_1187del"
   },
   {
      "transcriptId":"NM_001354370.2",
      "variantEffects":[
         "disruptive_inframe_deletion"
      ],
      "isCoding":true,
      "hgvsProtein":"p.(Y306_K307delins*)",
      "hgvsNucleotides":"c.918_920del"
   },
   {
      "transcriptId":"NM_015850.4",
      "variantEffects":[
         "disruptive_inframe_deletion"
      ],
      "isCoding":true,
      "hgvsProtein":"p.(Y395_K396delins*)",
      "hgvsNucleotides":"c.1185_1187del"
   },
   {
      "transcriptId":"NM_023105.3",
      "variantEffects":[
         "disruptive_inframe_deletion"
      ],
      "isCoding":true,
      "hgvsProtein":"p.(Y308_K309delins*)",
      "hgvsNucleotides":"c.924_926del"
   },
   {
      "transcriptId":"NM_023106.3",
      "variantEffects":[
         "disruptive_inframe_deletion"
      ],
      "isCoding":true,
      "hgvsProtein":"p.(Y306_K307delins*)",
      "hgvsNucleotides":"c.918_920del"
   },
   {
      "transcriptId":"NM_023110.3",
      "variantEffects":[
         "disruptive_inframe_deletion"
      ],
      "isCoding":true,
      "hgvsProtein":"p.(Y397_K398delins*)",
      "hgvsNucleotides":"c.1191_1193del"
   }
]

VEP results:

[steeljungle]

It seems something odd in VEP itself:
1:g.23688489_23688499delAATAAGAGCTGinsTT
1:g.23688489_23688499delAATAAGAGCTGinsT
These two variants actually all result in S459*, but VEP calls the first one as stop gained, but the second as frameshift.

[dmb107]

@steeljungle could you compile like 100 nonsense indels for me to test? I think we need more examples to figure out exactly how we want to fix.