CANCELLED: Seminar "Causation without correlation: Discovering nonintuitive fundamental relationships in transcriptional networks, neuroscience, and beyond" - by Dr. Gerald Pao

Date

Wednesday, February 19, 2020 - 14:00 to 15:00

Location

B503, Center Bldg.

Description

This seminar has been canceled.

 

-------------------------------------------------------------------------------------------------------------------------

Speaker: Gerald Pao, Ph.D., Salk Institute

 

Title: "Causation without correlation: Discovering nonintuitive fundamental relationships in transcriptional networks, neuroscience, and beyond"

 

Abstract:

Causation without correlation is a feature of complex systems but it is not commonly recognized although the fact that correlation does not imply causation is widely known. Here we show that causation without correlation is a fundamental feature of nonlinear complex dynamical networks. In our work we experimentally demonstrate the existence of causation without correlation in gene expression networks. Furthermore we show that the relationships of these systems can be mathematically discovered on the surfaces of low dimensional manifolds. These type of relationships also exist in neuronal dynamics and we show such examples from whole brain live imaging of zebrafish brains at single neuron resolution. Using this framework grounded in nonlinear dynamics, we show how this approach can be scaled to whole brains using the ABCI supercomputer. The goal of this work is to ultimately to “download” brains of real organisms into computers that control biosimilar robots. We also show some of the technologies we have generated while striving towards this goal such as manipulating the optical refractive index in living cells using cephalopod genes, as well as algorithm development that have sped up computation >5000 fold on supercomputers. 

-------------------------------------------------------------------------------------------------------------------------

All-OIST Category: 

Import this Event: Right-click to download, then open in your calendar application.

Subscribe to the OIST Calendar: Right-click to download, then open in your calendar application.