Marc Corrales

(Bio) Scientist/Researcher.

code / data / experiments.

Hi! thanks for visiting my corner.

My name is Marc Corrales, a (Bio) Researcher born in a mediterranean rock called Mallorca
My interests are varied but they revolve around understanding how matter has been able to organize in systems capable of encoding information to implement conditional outcomes, i.e how matter achieved control flow.
Examples of fields that draw my attention are: the theories of the origin/s of life, cellular “decision making”, evolution, game theory, epigenetics, gene regulation and specially neuroscience.
I am also attracted to the interface of computer science, data analysis and machine learning as ways to understand and model these phenomena.



I did my phD in Guillaume Filion’s lab at the Center for Genomic Regulation (2013-2018).

The aim of my thesis (Text in context: Chromatin effects in gene regulation) was to understand the role of chromatin context in gene expression. When I started my thesis, several recently published projects showed that despite the theoretical combinatorial complexity of chromatin factors (i.e. proteins, histone modifications), only a few discrete combinations were observed. Moreover these “chromatin states” were correlated with different gene expression outcomes.

As is frequently the case in systems with very correlated features the question was if the observed correlations were causal to gene expression and why there appeared to be different mechanisms to activate and silence transcription. To investigate these questions we generated barcoded reporter libraries that we randomly integrated in the genome allowing us to measure the chromatin context effect at ~ 85.000 different loci.

The surprise came when we set to predict the reporter expression with the vast amount of chromatin feature maps (~ 120) in addition to the three dimensional pattern of contacts. The expression of our housekeeping integrated reporters was greater when they landed in genomics regions contacting other active promoters and terminators: the predicton power was 2X better than when using enhancers or the chromatin composition features alone or in combination.

This data together with the endogenous organization of this type of genes allowed us to propose the following interpretation for the organization of the Drosophila genome: If frequent contacts with the terminators and promoters of active genes increase expression, housekeeping genes may benefit from being in spatial proximity to other active genes.In contrast, developmentally regulated genes should be shielded from transcriptional interference; the same principles thus explain why they are typically long and isolated.

Such an organization has the benefit of maintaining activators of transcription close to active genes, thereby reducing accidental activation of other genes.

We published our results in Genome Research (go to publications) to read the paper.

Recent posts

Recent tweets