cansavvy/microglia-multi-plier

Exploring microglia activation gene expression signatures using multi-plier techniques and refine.bio data.

Microglia activation gene expression signature identification

Scientific Objectives:

Part one) Identify latent variables indicative of different types of microglial activation using the unsupervised learning transfer techniques used in MultiPLIER

Part two) Use identified microglial activation gene signatures to evaluate microglia activation in various disease contexts.

Project maintenance objectives:

• All code version controlled on GitHub
• Maintain Google R code style: http://web.stanford.edu/class/cs109l/unrestricted/resources/google-style.html
• Pull Request model of code review

Methodological steps:

Data Collection

Collect pre-normalized gene expression datasets from refine.bio that meet these criteria previously determined:

1. Gene expression has to have representation of coding transcripts, as opposed to specialty assays that only assay non-coding transcripts or a predetermined set of transcripts.
2. Studies must include human microglia sample data.
3. Studied tissue must be a cell line, or from brain, blood, or CSF.
4. Gene expression assays must be done on isolated microglia as opposed to heterogenous cell-type samples.
5. Samples cannot be from a pathological state organism or have any other treatment done to them besides the microglial activation treatment.

Identification of gene expression signatures (latent variables)

Use MultiPLIER Code here: https://github.com/greenelab/multi-plier to identify latent variables associated with Microglia activation.

Writing up results

Use Manubot software to collaboratively write manuscript.