Skip to content

Commit fe54894

Browse files
kMutagenekMutagene
committed
docs: establish active ArcIR roadmap
1 parent ac74a78 commit fe54894

14 files changed

Lines changed: 434 additions & 0 deletions

‎plans/claude/arcir-conversion-restructure.md‎

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,7 @@
11
# Per-entity conversion restructure (BioProject exemplar)
22

3+
> **Status: DONE (2026-08-20).** The original acceptance criteria are implemented. This document is retained as historical design context; future evolution is governed by [the active implementation plan](../implementation.md).
4+
35
## Goal
46

57
Give each accession type its own conversion, authored explicitly so **object integrity is preserved** —

‎plans/claude/arcir-export-readiness.md‎

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,7 @@
11
# ArcIR → AI-readiness export plan (F2)
22

3+
> **Status: SUPERSEDED (2026-08-20).** R1 and R2 were implemented by the crawler, but the proposed R0/R3/R4 export architecture is incomplete and no longer authoritative. Implemented portions remain historical context; replacement work is consolidated in [the active implementation plan](../implementation.md).
4+
35
> Canonical plan. Merges the original shape doc with the 2026-07-10 design-session refinements
46
> (`plans/glm-5.2/arcir-export-readiness.md`): the locked decisions are authoritative, the name-helper
57
> implementations and tests are grounded against the committed fixtures, and the *rationale* behind each

‎plans/claude/arcir-graphml.md‎

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,7 @@
11
# ArcIR → GraphML visualization plan
22

3+
> **Status: DONE (2026-08-20).** The original GraphML and embedded HTML-viewer acceptance criteria are implemented. This document is retained as historical design context; replacement of the barebones HTML viewer and future visualization work are governed by [the active implementation plan](../implementation.md).
4+
35
## Overview
46

57
The INSDC → ArcIR mapping (`plans/arcir-mapping.md`) assembles a **property graph** —

‎plans/claude/arcir-ingest.md‎

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,7 @@
11
# Ingesting supplementary sources (paper + count data) into the ArcIR
22

3+
> **Status: DONE (2026-08-20).** The original acceptance criteria are implemented. This document is retained as historical design context; future evolution is governed by [the active implementation plan](../implementation.md).
4+
35
## Context
46

57
The forward mapping turns INSDC records into the ArcIR property graph (`Arc.Build.ArcIR` — typed

‎plans/claude/arcir-mapping.md‎

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,7 @@
11
# INSDC → ArcIR mapping plan
22

3+
> **Status: SUPERSEDED (2026-08-20).** The graph foundation was implemented, but the hybrid structural-ontology overlay described here is no longer authoritative. Implemented portions remain historical context; replacement work is consolidated in [the active implementation plan](../implementation.md).
4+
35
## Overview
46

57
`BioFSharp.INSDC.ArcIR` maps parsed INSDC records into **ArcIR** — an ARC-oriented intermediate

‎plans/claude/crawler.md‎

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,7 @@
11
# BioFSharp.INSDC.Crawler implementation plan
22

3+
> **Status: DONE (2026-08-20).** The original acceptance criteria are implemented. This document is retained as historical design context; future evolution is governed by [the active implementation plan](../implementation.md).
4+
35
## Context
46

57
The repo can already read/write and model INSDC records, but every record it has

‎plans/claude/implementation.md‎

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,7 @@
11
# BioFSharp.INSDC implementation plan
22

3+
> **Status: DONE (2026-08-20).** The original acceptance criteria are implemented. This document is retained as historical design context; future evolution is governed by [the active implementation plan](../implementation.md).
4+
35
## Overview
46

57
This repo ships two packages that together give BioFSharp users an INSDC (International Nucleotide Sequence Database Collaboration) read/write surface:

‎plans/claude/r1-crawlers.md‎

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,7 @@
11
# R1 crawlers — fetch the raw artifacts (DEE2 archive + paper)
22

3+
> **Status: DONE (2026-08-20).** The original acceptance criteria are implemented. This document is retained as an as-built historical record; future evolution is governed by [the active implementation plan](../implementation.md).
4+
35
> The ArcIR export plan (`plans/claude/arcir-export-readiness.md`) defines **R1** — the
46
> *un/weakly structured* AI readiness level format — as the DEE2 count archive verbatim plus the paper,
57
> with the source *obscured* (no INSDC XML). Like R2, **R1 needs no IR at all**: it just needs two raw

‎plans/claude/r2-crawlers.md‎

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,7 @@
11
# R2 crawlers — fetch the raw artifacts (INSDC XML, paper, DEE2 counts)
22

3+
> **Status: DONE (2026-08-20).** The original acceptance criteria are implemented. This document is retained as an as-built historical record; future evolution is governed by [the active implementation plan](../implementation.md).
4+
35
> The ArcIR export plan (`plans/claude/arcir-export-readiness.md`) defines R2 as "the full INSDC XML
46
> record tree in a folder layout + counts (+ paper)". R2 needs no IR at all — it just needs the three
57
> raw source artifacts on disk in a canonical tree. This plan covers the **source-side** crawlers that

‎plans/claude/sqlite-store.md‎

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,7 @@
11
# Plan: BioFSharp.INSDC.SQLite — Schema + Deconstruction/Reconstruction Layer
22

3+
> **Status: SUPERSEDED (2026-08-20).** The core store was implemented, but its focused acceptance suite and migration work remain incomplete. Implemented portions remain historical context; outstanding work is consolidated in [the active implementation plan](../implementation.md).
4+
35
## Context
46

57
The next phase of `BioFSharp.INSDC` is a SQLite-backed store for the INSDC entity hierarchy (`BioProject > Study > BioSample > Experiment > Run`), intended to feed a crawler for these metadata records.

0 commit comments

Comments
 (0)