Census enables users to perform in silico experimentation to interrogate genes responsible for cell states recorded in the CELLxGENE metadata (e.g. cell type, disease, development stage, etc) using pre-trained models to enable biologists to quickly test hypotheses in an experiment-free fashion.
The work defined here is for the Discovery phase of this project. A product phase should be described elsewhere and once this discovery phase is finalized.
User stories
I am a computational biologist who wishes to understand what genes have the highest potential to:
Go from a cancer state to a healthy state.
Go from a differentiated state to pluripotent state.
Go from differentiate state to another differentiated state (e.g. B cell to to T cell)
Goals
To have a proposed testes method for in silico experimentation to be considered for production.
Discovery work
The deliverables are the following:
Landscape current methods.
Draft product requirements and validation strategies.
Assess feasibility or re-use of current models/embeddings.
Definition
Census enables users to perform in silico experimentation to interrogate genes responsible for cell states recorded in the CELLxGENE metadata (e.g. cell type, disease, development stage, etc) using pre-trained models to enable biologists to quickly test hypotheses in an experiment-free fashion.
The work defined here is for the Discovery phase of this project. A product phase should be described elsewhere and once this discovery phase is finalized.
User stories
Goals
To have a proposed testes method for in silico experimentation to be considered for production.
Discovery work
The deliverables are the following: