Skip to content

Osteogenesis Imperfecta Type VII phenotype curation manual 2026 04 24

Osteogenesis Imperfecta Type VII phenotype curation notes

Date: 2026-04-19 Target file: kb/disorders/Osteogenesis_Imperfecta_Type_VII.yaml Scope: phenotype section only

Primary phenotype sources reviewed

  • PMID:12110406 - founding eight-person Quebec cohort; strongest source for fractures at birth, bluish sclerae, lower-extremity deformity, coxa vara, osteopenia, and rhizomelia.
  • PMID:19895918 - human OI type VII childhood cohort; supports impaired growth.
  • PMID:18566967 - recessive CRTAP/LEPRE1 OI cohort; supports extremely low bone mineral density and "popcorn" epiphyses in the severe neonatal/surviving spectrum.
  • PMID:38214665 - 2024 CRTAP paper; strongest recent source for early-onset recurrent fractures, severe osteoporosis, and bone deformities.
  • PMID:35970273 - focused craniofacial study in two girls with CRTAP-related OI; supports platybasia and wormian bones.
  • PMID:41064055 - 2025 clinical-variability paper; supports prenatal-vs-postnatal fracture variability and a cautious ocular expansion (high myopia with bilateral retinal detachment in one patient).

Key phenotype decisions

  • Softened Severe short stature to Growth Delay because the cited evidence supported impaired growth, not a uniform severe-short-stature claim.
  • Removed unsupported white sclerae wording from the top description; retained Blue sclerae only because the founding abstract explicitly reports bluish sclerae.
  • Added phenotype entries for osteoporosis, skeletal deformities, popcorn epiphyses, wormian bones, platybasia, high myopia, and retinal detachment.
  • Kept ocular findings explicitly cautious because the 2025 paper reports them in one patient rather than as a core defining feature.

Claims deliberately not added

  • Dentinogenesis imperfecta exclusion: the 2022 craniofacial abstract reports absence in two girls only, which is too narrow to encode as a disorder-level excluded phenotype.
  • Class III malocclusion / crossbite: clinically real in the 2022 abstract, but I did not add them because I could not quickly establish a clean, specific HPO mapping with the same confidence as the retained terms.
  • Narrow chest, crumpled long bones, white/light-blue sclerae, and similar lethal-neonatal details from figures/full text were left out because this pass was constrained to abstract-validated PMID evidence.