docs: add usage, output, and samplesheet schema documentation

[hyphaltip]

Summary

Complete the nf-core-inspired documentation and schema requirements (issue #12). Add comprehensive user-facing documentation for pipeline usage and output structure, along with a JSON schema for samplesheet validation.

Key changes

• docs/usage.md — Quick-start guide covering:

  • Profile composition (pipeline × executor × provisioning)
  • Samplesheet format with required/optional columns and examples
  • Runtime sample filtering (--asmid, --taxon, --n_test)
  • Key parameters with descriptions
  • Resume, stub-run, and EarlGrey masking workflows

• docs/output.md — Complete output directory reference:

  • Directory tree showing all output locations and purposes
  • Key outputs table (annotations, cleaned genomes, RNA-seq, statistics, TE libraries)
  • Explanation of the storeDir persistence model (outputs accumulate across runs; deleting a file forces re-execution)
  • Distinction between storeDir (most outputs) and publishDir (SRA, antiSMASH, InterProScan, SignalP)

• assets/schema_input.json — JSON Schema (draft 2020-12) for samplesheet validation:

  • Required fields: SPECIES, ASMID, LOCUSTAG, BUSCO_LINEAGE
  • Optional fields: STRAIN, TRANSL_TABLE, NCBI_TAXONID, GENOME, taxonomic ranks
  • Descriptions and error messages for each field
  • Defaults for optional fields

• nextflow.config — Add validation block:

  • lenientMode = true to accept string representations ("true"/"1") from CLI args in NF 26.x
  • monochromeLogs = false for readable schema validation output

• REFACTORING_PLAN.md — Update status:

  • Mark issues Funannotate predict subworkflow (Phase 7) #11 and Annotation subworkflow (Phase 8) #12 as complete
  • Record naming decision: stay nf-core-inspired (not submitted); keep nf_funannotate1 naming; defer container-per-module and nf-test as non-priorities for this HPC-first pipeline

• .github/REFACTOR_ISSUES.md — Close issue Annotation subworkflow (Phase 8) #12 checklist:

  • All documentation and schema items marked done
  • Naming + submission decision recorded

Implementation notes

• The samplesheet schema is referenced by nf-schema plugin (already in nextflow.config) for runtime validation
• Output documentation emphasizes the storeDir caching model, which is central to the pipeline's resume behavior on NFS-mounted filesystems
• Usage guide includes both common workflows (annotate, EarlGrey) and advanced filtering patterns
• All paths are relative to launchDir unless noted as overridable parameters

https://claude.ai/code/session_01AbWwmm9v6dJn13sgAJn3Sa