Replace existing download helpers¶
Use the library examples for the API calls. Migrate acquisition separately from changes to reference releases, allele selection, or analysis settings.
Replace one operation at a time¶
Install osteosarc normally: pysam and datacache are standard dependencies.
The older [reads] extra remains accepted for compatibility. The CLI now
prints a sample overview by default; use samples --json for its original
source-attributed claims.
| Existing code | Replacement |
|---|---|
| Download website metadata and save checksums | Dataset.sync(name) |
| Reopen pinned metadata | Dataset.open(name) |
| Parse the variant page and join count-export alleles | data.variants() |
| Find files by assay, timepoint, or path | data.assets.select(...) |
| Download and verify a whole file | data.download(asset) |
| Survey an alignment's reference | data.inspect_alignment(asset) |
| Fetch indexed regions and optional paired mates | data.extract_reads(asset, regions, ...) |
| Construct native Varcode alleles | variants.to_varcode(genome=...) |
Preserve the analysis¶
Keep reference releases, transcript selection, structural-variant definitions,
read filters, and scoring settings in the downstream project. Declare the
assembly for custom-named references with to_varcode(..., assembly="GRCh38").
Before retiring an old extractor, compare complete SAM records, including tags and the number of occurrences of each record. Equal read counts alone do not show that the records match.
subset_templates samples without using allele support or quality. It does not
reproduce Isovar's alternate-read enrichment or another project's fixture
selection policy. Use fixture recipes to freeze evidence assignments,
exact record multiplicity, controls and assembly context explicitly.
Review corrections¶
Osteosarc applies source corrections by default, including the
MAP2 allele change and five relocated Tempus alleles. Open a snapshot with
corrections=False when comparing against original published inputs. Review
changes to biological expectations separately from the acquisition migration.
Reuse old downloads¶
Use the shared OPENVAX_DATA_CACHE directory where possible. Import existing
files with Cache.import_file(path, original_url, sha256=...); see
snapshots and cache.
Keep the old manifests for their acquisition dates and provenance.