The National Institutes of Health awarded Northwestern's Center for Physical Genomics and Engineering a prestigious T32 grant to establish the Physical Genomics Training Program (PGTP), which provides stipend, tuition and other support for PhD students who are accepted into the program. The PGTP is the first physical genomics-based graduate training program in the United States.

This online seminar will feature 15-minute talks from three of the program's current PhD trainees, who will provide an overview of the cross-disciplinary research projects they are working on during their graduate study. Each talk will have a few minutes for Q&A afterward. Registration is free, but required at: https://tinyurl.com/yj5v8y85

Friday, July 25, 2025 @ 12pm CT - Live on Zoom

The Impact of Charge Regulation and Ionic Intranuclear Environment on the Nucleosome Core Particle

Paola Carrillo Gonzalez PhD Student, Backman & Szleifer Labs

We present a molecular theory to examine how the intranuclear environment regulates the charge of a nucleosome core particle (NCP), accounting for the size, shape, charge, and chemical state of all components. By varying monovalent and divalent salt concentrations and pH, we quantify how charge regulation arises from amino acid and DNA-phosphate dissociation equilibria, electrostatic interactions, and mobile ion entropy. Notably, divalent magnesium ions strongly modulate NCP charge via ion bridging, leading to charge neutralization and even inversion, consistent with experimental observations. Our results highlight the necessity of incorporating detailed molecular interactions, such as DNA-phosphate ion condensation and histone acid–base chemistry, to accurately predict NCP charge states.

Deep Learning Modeling of RNA Translation Initiation in Human Cells and Diseases

John Carinato PhD Student, Ji & Backman Labs

The dysregulation of gene expression is the cause of many human diseases, including cancers. However, the genomics field lacks systematic methods to quantitatively analyze translation initiation. Due to their enhanced flexibility, deep neural networks have the capability to identify all cis-regulatory elements involved in controlling the process of translation initiation. Here, we built multiple deep learning models with advanced architectural units that have produced highly accurate predictions of translation initiation sites across the genome. Deep learning has afforded us the ability to quantify the impact an AUG codon, the canonical start codon, has on translation initiation compared to one of its single nucleotide variants. Furthermore, we can predict all locations of canonical and non-canonical translation initiation across human transcripts, based solely on a transcript’s genetic sequence. Determining where translation initiation occurs in the genome will allow for a comprehensive translation initiation landscape, and thus gene expression profile, to be understood in human diseases such as cancers.

Exploring the Role of Heterochromatin Protein 1 Alpha in Stabilizing Chromatin Packing Domains

Tiffany Kuo PhD Student, Backman Lab

The human genome is organized in a hierarchical manner to facilitate gene expression and overall genome function. Prior work has shown the existence of chromatin packing domains that demonstrate mass scaling behavior. Transcription and cohesin/CTCF-mediated loop formation create nascent domains. The size of nucleosome remodeling enzymes plays a major role in allowing heterochromatin enzymes to more easily penetrate areas of high density in the cores of domains to facilitate maturation. Furthermore, transcription has a nonmonotonic dependence of RNA polymerase on local crowding conditions and happens in optimal conditions. By using live-cell Dual Partial Wave Spectroscopic (Dual-PWS) microscopy and Stochastic Optical Reconstruction Microscopy (STORM), this study aims to uncover how heterochromatin protein 1 alpha (HP1α), a critical chromatin crosslinking protein that binds to H3K9me3, is involved in stabilizing chromatin packing domains. By exploring how HP1α is involved in domain formation and maturation, we hope to manipulate transcriptional memory in various disease contexts, such as aging and cancer.

Supported by NIH grants T32GM142604 and U54CA268084

Related Info

Register Now