Research

RNA Networks in Disease

As my postdoc comes to an end, I plan to study various modes of RNA regulation so that we can use RNA measurements to understand dysregulation in disease.

Given the vast amount of information encoded in RNA it is important to understand how changes in RNA regulation such as alternative splicing, non-coding RNA interactions, alternative poly-adenylation and of course transcription can have broader effects on gene expression.

My postdoctoral research uncovered dramatic regulation of transcription by microRNAs, suggesting that post-transcriptional and transcriptional regulation of RNA are intrinsically linked.  Going forward, I plan to determine how other modes of RNA regulation interact and use this knowledge to build better computational models of RNA networks in disease.


 

To what extent do microRNAs (miRNAs) affect gene expression?

miRNAs are small non-coding RNAs that have been shown to drive developmental transitions and diseases such as cancer but cause only modest repression of their direct targets.  I hypothesized that miRNAs regulate transcription factors to affect global gene expression changes, and tested this hypothesis using numerous high-throughput datasets collected from cells with and without Dicer, an enzyme required for miRNA expression as shown below.

miRNA-transcriptional strategy
miRNA network summary: (A) Hypothesis of miRNA influence on transcription factors. (B) Graphical interpretation of network hypothesis. (C) High throughput data sets collected and analyzed to identify transcription factors of interest.

This graphical approach uncovered a large transcriptional network regulated by microRNAs, and can be used to study various other questions surrounding RNA regulation:

  • How do miRNA regulate gene expression through the protein-protein interaction network?
  • How much does feedback between miRNAs and transcription factors alter gene expression?
  • How can miRNAs impact alternative splicing?

 


Building better models of RNA regulation

RNA levels can be altered by transcriptional and post-transcriptional regulation, as shown below.

: Various modes of RNA regulation are not independent: transcriptional (grey box) and post-transcriptional (colored boxes) regulation of RNA alters expression of protein (red). Translated proteins can then alter RNA levels (colored arrows) through binding with DNA (grey box) or RNA to cause alternative poly-adenylation (blue box), alternative splicing (purple box), microRNA-mediated repression (yellow box) and interactions with long non-coding RNAs (orange box).  As such, each mode of RNA regulation can impact another.
Various modes of RNA regulation are not independent: transcriptional (grey box) and post-transcriptional (colored boxes) regulation of RNA alters expression of protein (red). Translated proteins can then alter RNA levels (colored arrows) through binding with DNA (grey box) or RNA to cause alternative poly-adenylation (blue box), alternative splicing (purple box), microRNA-mediated repression (yellow box) and interactions with long non-coding RNAs (orange box). As such, each mode of RNA regulation can impact another.

Given the substantial impact of miRNA regulation on mRNA via transcription, it is necessary individually study all types of RNA regulation shown above to better understand identify any links and use this knowledge to develop better network models.  Specifically I want to ask:

  • How does alternative splicing effect transcriptional regulation?
  • How do miRNAs impact alternative splicing?
  • How does alternative poly-adenylation impact the other modes of regulation?
  • To what extent does non-coding RNA play a role in gene expression?


Building better computational algorithms with RNA

Despite the variety of high throughput technologies available to measure RNA, DNA and binding assays such as ChIP and CLIP, no single assay can fully quantify the signaling dysregulation that arises in human disease. For example, proteins with altered function may not be differentially expressed at the transcript level. As such, it is important to integrate numerous data types to identify functional changes in a given phenotype.
Graphical models are a flexible way to integrate these data using measurements of RNA changes within a population (e.g. patients, single cell measurements and tissue types) and related epigenetic measurements. For example, knowing that miRNAs indirectly regulate mRNAs through transcription factors, I can construct networks from patient miRNA and mRNA data with related epigenetic data that link miRNAs associated with patient prognosis to mRNAs. The inferred network will contain proteins that are likely altered in disease that are not detectable at the RNA level. The more I learn about RNAs, the more types of RNA regulation I can include in the networks. Due to the ease with which RNA levels can be measured in the cell, this approach can be applied in a variety of biological contexts.

For example, we can use better models of RNA to ask the following questions:

  • Using miRNA and mRNA levels across patient cohorts in The Cancer Genome Atlas (TCGA), how can we identify putative drug targets of interest?
  • Using miRNA and mRNA levels together with epigenetic data from the NIH Roadmap Epigenomics Project what proteins are most significant in distinguishing tissues?
  • Using single-cell sequencing measurements, how can we identify specific proteins that give rise to differences in mRNA levels?

One thought on “Research

  1. I like your research questions very much, especially two questions regarding:
    How do miRNAs impact alternative splicing?
    How does alternative poly-adenylation impact the other modes of regulation?
    Could you please elaborate it little bite more how you gonna address these questions by computational modeling.

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s