Reconstructing gene regulatory networks with neural relational inference (GENERIC)

Discovering structures and dynamics of gene regulatory networks (GRNs)

The project at a glance

Start date:

01 Jul 2022
Duration in months:

48
Funding:

University of Luxembourg
Principal Investigator(s):

About

Discovering structures and dynamics of gene regulatory networks (GRNs) based on large-scale high throughput data is a fundamental research challenge in systems biology. The reconstruction of accurate GRNs from data is challenging because of not only the large number of parameters and components involved, but also their possible mutual interactions. The inferred network models can provide system-level understanding of the mechanism of cell transformations, which is crucial for the comprehension of the progression of complex diseases such as cancer. GENERIC takes an interdisciplinary approach towards GRN inference, integrating techniques from biology and computer science. We aim to devise data-driven deep neural models for inferring GRNs with high accuracy from single-cell gene expression data. In particular, we will be the first to develop novel GRN inference methods by exploiting neural relational inference (NRI) to discover latent interaction structures directly from data. This paves a new way for scalable reconstruction of real-size GRN models. By collecting, pseudo-time ordering, and processing large single-cell genomic datasets, GENERIC will develop NRI-based methods for effective identification of Boolean networks (BNs) as models of large-size GRNs. BNs are simple and yet are able to capture the important dynamic properties of the GRN under study. Our methods combine an iterative graph generation process and reinforcement learning into NRI to increase its efficiency and accuracy, and as well to incorporate prior biological knowledge and well-established network structural and dynamic properties. The newly developed GRN inference methods will be implemented as an open-source software tool. GENERIC will make an important breakthrough in-network inference, allowing more accurate and comprehensive GRN models inferred from single cell gene expression data.