CCB Brown Bag Seminar

1st Speaker: Maria Avdeeva, Ph.D., Associate Research Scientist, Developmental Dynamics

Topic: Quantifying effects of signaling perturbations with single cell transcriptomics

Integration of single cell measurements is a common task that aims to identify and relate cell states across individuals, treatments, or modalities. We adapt a previously developed integrative non-negative matrix factorization (iNMF) framework to a new setting where we simultaneously correct for batch effects and integrate between treatment conditions. We demonstrate the utility of our approach on simulated single cell transcriptomic data and apply it to newly collected scRNAseq datasets, to describe the effects of abnormal extracellular signal-regulated kinase (ERK) signal transduction in the early zebrafish embryo. In the wildtype embryo, the ERK pathway drives differentiating cells towards unique gene expression states. We characterize diverse gene expression responses to ERK activation by integrating transcriptomic profiles of wildtype zebrafish embryos and those that express an ERK-activating tool, photoswitchable MEK (psMEK). Using the iNMF framework, we describe changes introduced by the treatment in terms of the cell loadings and treatment-specific gene factors. We find that the treatment does not only redistribute the wildtype cell states, but also introduces an aberrant cell state which is not observed in the wildtype at this stage of development.

2nd Speaker: Miro Astore, Ph.D., Flatiron Research Fellow, SMBp

Topic: Can you see an ion channel move by freezing it?

Single particle cryogenic electron microscopy (cryoEM) involves the imaging of biomolecules trapped in vitreous ice. This method has revolutionized our understanding of cellular components because of its ability to deliver high resolution snapshots of biomolecules. However, conventional analysis of these datasets focuses on the construction of static structures, discarding any information about dynamics. This heterogeneity is critical to many biological processes such as the sensation of temperature.

Here we will demonstrate the use of machine learning based techniques to study heterogeneity in a cryoEM dataset for a temperature sensitive ion channel.This channel, TRPV1, is sensitive to both heat and capsaicin and is being pursued as a novel drug target for the treatment of chronic pain. Despite its importance, many aspects of its function including temperature sensation are poorly understood. It is hoped that a combination of molecular simulation and cryoEM experiments will assist us to tease apart this intriguing physical mechanism.