Fix endogenous genomic mapping #68
Conversation
This change reduces likelihood of errors when mapping accession-based transcript variants to genomic positions. Keep alignment process for now so that we can report the transcript and gene for these variants. Eventually, just the transcript id could be used to access this information.
bencap
left a comment
bencap
left a comment
There was a problem hiding this comment.
Looks good, thanks Sally. A couple minor comments that should be quick but which I do think are worth addressing before merging.
|
|
||
| _logger = logging.getLogger(__name__) | ||
|
|
||
| CLINGEN_API_URL = "https://reg.genome.network/allele" |
There was a problem hiding this comment.
Let's set this as an environment variable.
This should be a separate issue, but this did make me think we should standardize the way we setup envvars for ClinGen (and our other services, generally)
| try: | ||
| response.raise_for_status() | ||
| except requests.HTTPError as e: | ||
| msg = f"Received HTTPError from {CLINGEN_API_URL} for HGVS {hgvs}" | ||
| _logger.error(msg) | ||
| raise ResourceAcquisitionError(msg) from e |
There was a problem hiding this comment.
We might want to back off certain response errors with some retry logic here so we can do our best to fill everything in completely and avoid missing data from transient errors.
| """Provide dcd mapping version""" | ||
|
|
||
| dcd_mapping_version = "2025.3.1" | ||
| dcd_mapping_version = "2026.1.0" |
There was a problem hiding this comment.
We can bump this just prior to being released rather than as part of any individual changes. It's pedantic, but it's not really a new version until we are ready to release it.
Use ClinGen to map accession-based transcript variants to genomic locations.