Change Log¶
This page lists changes for the Beacon+ implementation of the "Beacon" genomics API, as well as related updates for the Progenetix front-end.
2024-11-02: Fix for missing TCGA analyses¶
During import of the TCGA non-CNV variants the "analysis" objects somehow got
lost although variants were imported correctly, leading to a missing association of
those variants w/ their samples during query result aggregation. Fixed for all
10007 samples by extracting the previously assigned analysis_id
values together
with the biosample_id
values from the variants and generating the missing analysis
objects. Now TCGA has working calls for "double hit" CNV + SNV events.
2024-09-10: Ongoing removal of samples from pure germline series¶
So far of the currently 145264 samples the Progenetix collection consisted of besides data from
- 112674 cancer samples (including primary tumors, recurrences and metastases)
- 5742 cancer cell line samples
- 26848 germline samples consisting of
- germline references of tumor patients
- 3201 reference profiles from the 1000 genomes project
- samples from 58 "accidental" germline series (i.e. such being analyzed in automated download & processing and then being kept for comparative purposes)
We are now in the process of creating a new resource for the germline samples which will become available in the near future under the refcnv.org address. While strengthening the focus of the Progenetix database this will provide a new general resource for germline CNV data e.g. as reference for rare disease applications.
2024-06-18 Switching ontology use for individual.sex
¶
While we had previously used the PATO terms for genotypic sex
(PATO:0020001
or PATO:0020002) we have now recoded those to "NCIT:C16576": "female" and "NCIT:C20197": "male"
to stay in line with the Beacon v2 documentation examples (which will probably drive implementations).
sex:
label: Genotypic Sex
infoText: |
Genotypic sex of the individual.
defaultValue: ""
isHidden: true
options:
- value: ""
label: "(no selection)"
- value: NCIT:C16576
label: female
- value: NCIT:C20197
label: male
2024-02-24 Adding analysis_operation
to analyses
¶
This new parameter with its (so far) values
"analysis_operation.id":"EDAM:operation_3961", "analysis_operation.label":"Copy number variation detection"
"analysis_operation.id":"EDAM:operation_3227", "analysis_operation.label":"Variant Calling"
... allows now the filtering of analyses based on the type of genomic profiling performed.
2023-06-12: Some schema changes...¶
Schema changes are now extensively tracked in bycon.progenetix.org
The latest changes of the database schemas involve e.g.:
- several field changes in biosamples, to align w/ main Beacon v2 default schema:
sampledTissue
=>sampleOriginDetail
description
=>notes
timeOfCollection.age
=>collectionMoment
2022-12-22: Various API fixes & extensions¶
- the
genespans
responses contain now acytobands
parameter (i.e. the cytogenetic mapping of the gene's locus) - FIXED: the beacon response meta now has the standard
receivedRequestSummary.requestParameters
instead of beforereceivedRequestSummary.variantPars
2022-01-17: Term-specific queries¶
So far (and still as standard), any
selected filter will also include matches on its child terms; i.e. "NCIT:C3052 -
Digestive System Neoplasm" will include results from gastric, esophagus, colon
... cancer. Here we introduce a selector for the search panel to make use of the Beacon v2
filters includeDescendantTerms
pragma, which can be set to false if one only
wants to query for the term itself and exclude any child terms from the matching.
Please be aware that this can only be applied globally and will affect all filtering terms used in a query. More information is available in the Filtering Terms documentation.
2022-01-17: Introducing variant_state
classes for CNVs¶
More information can be found in the description of ontology use for CNVs.
2022-01-10: BUG FIX Frequency Maps¶
Pre-computed Progenetix CNV frequency histograms (e.g. for NCIT codes) are based
samples from all child terms; e.g. NCIT:C3262
will display an overview of all
neoplasias, although no single case has this specific code.
However, there had been a bu when under specific circumstances (code has some mapped samples and code has more samples in child terms) only the direct matches were used to compute the frequencies although the full number of samples was indicated in the plot legend. FIXED.