Migrating from Genomics API v1beta2 to v1¶
The v1beta2 version of the Google Genomics API is deprecated.
If you use the the Google Genomics v1beta2 API, you will need to update your source code to use the v1 API.
Most changes are minor and updating your code will simply involve changing the name of objects or methods.
API Documentation¶
New client libraries¶
Python, Java, Go, and other language-specific client libraries for the v1 API can be found at:
API Changes¶
New endpoint:¶
- v1beta2:
https://www.googleapis.com/genomics/v1beta2 - v1 :
https://genomics.googleapis.com/v1
End of paging response¶
When reaching the end of paged-responses, the v1beta2 API would omit
the nextPageToken field from the response. The v1 API will always
return the nextPageToken field. When end of paging is reached,
the value will be an empty string.
New names¶
| v1beta2 | v1 |
|---|---|
| variantsets.importVariants | variants.import |
| variantsets.mergeVariants | variant.merge |
| jobs.cancel | operations.cancel |
| jobs.get | operations.get |
| jobs.search | operations.list |
Changed parameters¶
For Google Cloud project references, the v1beta2 API used the Cloud project number, while the v1 API uses the Cloud project id.
This change impacts how you call APIs. The following APIs have changed from accepting an input project number to a project id:
- datasets.list
- jobs.search (now operations.list)
- readgroupsets.export
- variantsets.export
This change impacts objects returned from APIs. The following objects have changed to contain a project id instead of a project number:
- datasets
- jobs (now operations)
New datatypes¶
| Field | v1beta2 | v1 |
|---|---|---|
| dataset.create_time | long | A timestamp in RFC3339 UTC “Zulu” format |
| read.info | map<string, array<string>> | map<string, array<object>> |
| annotations.info | map<string, array<string>> | map<string, array<object>> |
New objects¶
Operations replaces jobs¶
The jobs object has been replaced with operations.
Job Status¶
With v1beta2, the status was encoded in the status field with
detailed information in the detailedStatus.
With v1, the status is a combination of status field plus the
existence of the either the errors object (failure) or the
response object (success).
ID¶
With v1beta2, the job was identified with the id field.
With v1, the operation is identified by the name field.
importedIds¶
After a successful import, the job object would be populated with a list of
importedIds.
In the operation, this data is a child field of the response, either the
callSetIds or readGroupSetIds.
created¶
The job created time is now in the operation metadata.createTime field.
request¶
The job request values are now in the operation metadata.request field.
Field Masks¶
All Genomics APIs accept an optional list of fields to return.
This is sometimes referred to as a “field mask”.
The patch and update APIs accept an updateMask indicating the specific
fields to change.
In v1beta2, the “/” character could be used to separate a parent object from a child field, for example:
readGroupSets/id
In v1, the “/” character is not accepted, and parentheses must wrap the child field:
readGroupSets(id)
You can test building a proper field mask by using the “fields editor” available in the documentation for the Google Genomics API, or from the Google APIs Explorer.
For example, for genomics.datasets.list see: