Add phenopacket extraction template

[caufieldjh]

No description provided.

[caufieldjh]

very initial results on a case report, without expanding the subcategories (from the text of PMID 10874631):

extracted_object:
  id: 0bc06850-bb1b-4ee4-914a-74df417ef5c0
  label: Mutations in the retinal specific ATP binding transporter gene (ABCR) related
    to retinal diseases
  subject: First cousin II.1; First cousin II.2
  phenotypic_features:
    - Bilateral loss of central vision
    - Gradual loss of visual acuity
    - Choriocapillaris atrophy
    - Night blindness
    - Retina pigmentary deposits
    - Macular dystrophy
    - Visual field showed a central scotoma
    - Photopically abnormal ERG
    - Progressive night blindness
    - Concentric reduction of visual field
    - Total blindness with visual acuity reduced to light perception
  measurements:
    - Visual acuity
    - Fundus appearance
    - Fluorescein angiography
    - Goldman perimetry
    - Electroretinogram (ERG)
  biosample:
    - Blood samples from the patients and family
  interpretations:
    - RP19 and STGD are allelic disorders at the ABCR locus despite clinical heterogeneity
  diseases:
    - Stargardt disease
    - Retinitis pigmentosa (RP19)
  medical_actions:
    - NA
  files:
    - NA
  meta_data: NA

[caufieldjh]

This is now readily usable.
Example with the following vignette from phenopacket2prompt:

[source]
pmid = PMID:16962354
title = Functional analysis of mutations in TGIF associated with holoprosencephaly
[diagnosis]
disease_id = OMIM:142946
disease_label = Holoprosencephaly 4
[text]
A male proband presenting with lobar holoprosencephaly, atypical ventricles with small frontal horns,
hypothalamic and caudate fusion, diabetes insipidus, seizures,
premaxillary agenesis, microcephaly, absent nasal root and septum with a
depressed nasal tip (Fig. 1A and B).

$ ontogpt -vvv extract -t phenopackets -i ~/phenopacket2prompt/docs/cases/PMID_16962354.txt -m gpt-4o

extracted_object:
  phenopackets:
    - id: Proband_1
      subject:
        sex: MALE
        taxonomy: NCBITaxon:9606
      phenotypic_features:
        - description: Lobar holoprosencephaly
          type: HP:0006870
        - description: atypical ventricles with small frontal horns
          type: AUTO:ventricles
        - description: Fusion observed between the hypothalamus and the caudate region.
          type: AUTO:Hypothalamic%20and%20caudate%20fusion.
        - description: Excessive thirst and excretion of large amounts of severely
            diluted urine, with a reduction of fluid intake having no effect on the
            concentration of the urine.
          type: HP:0000873
        - description: seizures
          type: HP:0001250
        - type: AUTO:premaxillary%20agenesis
        - type: HP:0000252
        - description: absent nasal root and septum with a depressed nasal tip
          type: HP:0003196
      diseases:
        - term: MONDO:0007734
      meta_data: Associated study includes analysis of mutations in TGIF linked to
        Holoprosencephaly 4 (PMID:16962354)
named_entities:
  - id: HP:0006870
    label: Lobar holoprosencephaly
    original_spans:
      - 224:246
  - id: AUTO:ventricles
    label: ventricles
    original_spans:
      - 258:267
  - id: AUTO:Hypothalamic%20and%20caudate%20fusion.
    label: Hypothalamic and caudate fusion.
  - id: HP:0000873
    label: Diabetes insipidus
    original_spans:
      - 328:345
  - id: HP:0001250
    label: seizures
    original_spans:
      - 348:355
  - id: AUTO:premaxillary%20agenesis
    label: premaxillary agenesis
    original_spans:
      - 358:378
  - id: HP:0000252
    label: microcephaly
    original_spans:
      - 381:392
  - id: HP:0003196
    label: nasal hypoplasia
  - id: MONDO:0007734
    label: Holoprosencephaly 4
    original_spans:
      - 166:184

[caufieldjh]

Will still need some tuning and fixes to improve extraction and grounding, of course.