Update Use Case description and API change to accept GCP auth token (#4384)

* pass auth_token in create_http_client

* lint on the notebook

* feat:Removed the last update date

* Update notebooks/community/alphagenome/README.md

Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

* Update notebooks/community/alphagenome/cloudai_alphagenome_vai_quickstart.ipynb

Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

---------

Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
This commit is contained in:
Damodar Panigrahi
2025-12-05 18:30:23 +00:00
committed by GitHub
co-authored by gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
parent 4ab197a4ba
commit 820c0f8db4
2 changed files with 17 additions and 25 deletions
+4 -20
View File
@@ -40,29 +40,13 @@ human and mouse cell types and tissues.
![Diagram showing an overview of the AlphaGenome model architecture and its inputs/outputs](https://www.alphagenomedocs.com/_images/model_overview.png)
## Use Cases
* **Predict outputs for a DNA sequence:** AlphaGenome is a model that makes
predictions from DNA sequences. AlphaGenome predicts multiple 'tracks' per
output type, covering a wide variety of tissues and cell-types.
* **Sequence-to-function predictions:** Predict multiple functional tracks (such as gene expression, splicing) from DNA sequences across a wide variety of tissues and cell types.
* **Open-vocabulary object retrieval:** AlphaGenome can make predictions for
a human reference genome sequence specified by a genomic interval.
For example, let's predict RNA-seq for tissue 'Right liver lobe' in a 1MB
region of Chromosome 19 around the gene CYP2B6, which encodes an enzyme
involved in drug metabolism, and is primarily expressed in the liver.
* **Variant effect scoring:** Assess the impact of genetic variants by comparing predictions for the reference and alternative alleles and summarising the differences between them.
* **Predict variant effects:** AlphaGenome can predict the effect of a
variant on a specific output type and tissue by making predictions for the
reference (REF) and alternative (ALT) allele sequences.
* **Identify functional regions:** Use in silico mutagenesis (ISM) to identify functionally important regions in the DNA sequence.
* **Scoring the effect of a genetic variant:** AlphaGenome can score the
effect of a genetic variant by making predictions for the REF and ALT
sequences and aggregating the track signal. To highlight which regions in a
DNA sequence are functionally important for a final variant prediction,
AlphaGenome can help you to perform an in silico mutagenesis (ISM) analysis
by scoring all possible single nucleotide variants in a specific interval.
* **Human and mouse predictions:** AlphaGenome can generate predictions for
both humans and mouse.
* **Human and mouse capability:** Generate predictions for both human and mouse genomes.
## Documentation
This API provides access to AlphaGenome, Google DeepMind's unifying model for
@@ -131,8 +131,12 @@
"source": [
"vertex_ai_url = \"\" # @param {type:\"string\"}\n",
"service_account = \"\" # @param {type:\"string\"}\n",
"code_whl = \"alphagenome-0.4.2-py3-none-any.whl\"\n",
"file_path = f\"gs://alphagenome-whl/{code_whl}\""
"token = \"\" # @param {type:\"string\"}\n",
"code_whl = \"alphagenome-0.4.2.2-py3-none-any.whl\"\n",
"file_path = f\"gs://alphagenome-whl/{code_whl}\"\n",
"\n",
"service_account = service_account or None\n",
"token = token or None\n"
]
},
{
@@ -268,9 +272,11 @@
"source": [
"print(\"Creating HttpDnaClient...\")\n",
"dna_model = dna_client_http.create_http_client(\n",
" vertex_ai_url=vertex_ai_url, model_version=\"FOLD_0\", service_account=service_account\n",
" vertex_ai_url=vertex_ai_url,\n",
" model_version=\"FOLD_0\",\n",
" service_account=service_account,\n",
" token=token,\n",
")\n",
"# dna_model = dna_client_http.create_http_client(vertex_ai_url=vertex_ai_url, model_version=\"FOLD_0\")\n",
"print(\"HttpDnaClient created.\")"
]
},
@@ -4616,7 +4622,9 @@
}
],
"source": [
"interval = Interval(chromosome=\"chr19\", start=2**20, end=2**20 + 2**20, strand=\".\")\n",
"interval = Interval(\n",
" chromosome=\"chr19\", start=2**20, end=2**20 + 2**20, strand=\".\"\n",
")\n",
"\n",
"scores = dna_model.score_interval(\n",
" interval,\n",