PQG Student and Postdoc Seminar

Each year, the PQG organizes a less formal PQG Student and Postdoc Seminar for all local students, postdocs, and faculty. The goal is to provide the opportunity to present and participate in the discussion of works-in-progress, and to focus on the methods and analysis of high-dimensional data in genetics and genomics.

2023/2024 Student and Postdoc Seminar organizers: Julia Sealock & Hui Li

Please direct any logistical questions to Amanda King

Note: Harvard Chan School seeks to bring in speakers with a wide range of experiences and perspectives. They’re here to share their own insights; they do not speak for the school or the university.

Upcoming Seminar

All PQG Student and Postdoc Seminar meetings for the semester will be held in person.

Tuesday, March 19, 2024
1:00-2:00 PM

Kodi Taraszka

Research Fellow in Medicine
Dana-Farber Cancer Institute

COX proportional hazards Mixed Model (COXMM) accurately estimates the heritability of time-to-event traits

With large biobanks connecting electronic health records with genetic sequencing, our understanding of the genetic architecture of time-to-event (TTE) traits such as age-of-onset, treatment response, and disease progression has grown. As a result, several genome-wide association study (GWAS) methods have been developed based on the TTE phenotypic generative model (PGM); however, all existing heritability methods still model a linear relationship between the trait and genetics. Here, we propose a new heritability method, COXMM, a COX proportional hazard Mixed Model designed to estimate the heritability of traits which follow a TTE PGM. We demonstrate the efficacy of COXMM for TTE heritability estimation, both in simulations and in the UK Biobank.

2023-2024 Dates

October 10, 2023 - Rongbin Zheng, Boston Children’s Hospital

Rongbin Zheng

Research Fellow in Pediatrics
Boston Children’s Hospital

Computational Models on Metabolite-mediated Intercellular Communication

We developed MEBOCOST, a computational algorithm for quantitatively inferring metabolite-based intercellular communications using single cell RNA-seq data. MEBOCOST predicted cell-cell communication events for which metabolites, such as lipids, are secreted by one cell (sender cells) and traveled to interact with sensor proteins of another cell (receiver cells). The sensor protein on receiver cell might be cell surface receptor, cell surface transporter, and nuclear receptor. MEBOCOST relies on a curated database of metabolite-sensor partners, which we collected from the literatures and other public sources. Based on scRNA-seq data, MEBOCOST identifies cell-cell metabolite-sensor communications between cell groups, in which metabolite enzymes and sensors were highly expressed in sender and receiver cells, respectively. Applying MEBOCOST on brown adipose tissue (BAT) showed the robustness of predicting known and novel metabolite-based autocrine and paracrine communications. Additionally, MEBOCOST identified a set of intercellular metabolite-sensor communications that was regulated by cold exposure in BAT. Those predicted communicating metabolites and sensors may play important roles in thermogenesis regulation. We believe that MEBOCOST will be useful to numerous researchers to investigate metabolite-based cell-cell communications in many biological and disease models, thus will be useful to remove critical barriers impeding the development of new therapies to target these communications. MEBOCOST is freely available at https://github.com/zhengrongbin/MEBOCOST.

November 14, 2023 - Eric Van Buren, HSPH

Eric Van Buren

Postdoctoral Fellow in Biostatistics & Statistical Genetics
Harvard T.H. Chan School of Public Health

cellSTAAR: Incorporating single-cell-sequencing-based functional data to boost power in rare variant association testing of non-coding regions

Whole genome sequencing studies have identified hundreds of millions of rare variants, the majority of which are in non-coding regions and of unknown function. Because the regulatory landscape of many candidate Cis-Regulatory Elements (cCREs) varies across cell types, it is of substantial interest to incorporate single-cell sequencing data into RVATs of cCREs to capture the functional variability that exists across cell types in the non-coding genome and boost statistical power in the process. We propose cellSTAAR to address two opportunities to improve existing gene-centric RVAT methods as applied to genetic variants in cCREs. First, cellSTAAR integrates single-cell ATAC-seq data to capture variability in chromatin accessibility across cell types via the construction of cell-type-specific variant sets and cell-type-specific functional annotations. Second, cellSTAAR links cCREs to their target genes using an omnibus framework that aggregates results from a variety of linking approaches to reflect the uncertainty in element-gene linking. We applied cellSTAAR on Freeze 8 (N = 60,000) of the NHLBI Trans-Omics for Precision Medicine (TOPMed) consortium whole genome sequencing data to three quantitative lipids traits: LDL, HDL, and TG. In at least one cell type, genome-wide significant promoter and enhancer associations were found in several known lipids loci, including APOE, APOA1, and CETP. Unlike existing methods, cellSTAAR reveals variability in the significance at these loci across a variety of cell types from diverse tissues and uncertainty in the target gene for significant enhancers. Using a weakened genome-wide significance threshold, the most discoveries using cellSTAAR are found in cell types that are the most relevant to lipids such hepatocytes. We replicate our results using UK Biobank whole genome sequcing data (N = 190,000).

December 12, 2023 - Kaia Mattioli, Brigham and Women's Hospital

Kaia Mattioli

Postdoctoral Research Fellow
Brigham and Women’s Hospital

Widespread variation in molecular interactions and regulatory properties among transcription factor isoforms

Transcription factors (TFs) control gene expression by interacting with DNA and cofactors to regulate transcription. Human TF genes produce multiple protein isoforms with altered DNA binding domains, effector domains, and other protein regions. The global extent to which this results in functional differences between TF isoforms of the same gene remains unknown. Here, we systematically tested 756 isoforms of 309 TF genes, assessing DNA binding, protein binding, transcriptional activation, subcellular localization, and condensate formation. Compared to the reference isoform, two-thirds of alternative TF isoforms exhibit differences in their molecular activities, which often cannot be predicted from sequence alone. We observed two primary categories of alternative TF isoforms: “rewirers” and “negative regulators”, both of which are associated with differentiation and cancer. Our results support a model wherein the relative expression levels of and interactions between TF isoforms add an understudied layer of complexity to gene regulatory networks, demonstrating the importance of isoform-aware characterization of TF functions and providing a rich resource for further studies.

Feb 13, 2024 - Jennifer Chen, Harvard

Jennifer Chen

Research Associate, Department of Molecular
and Cellular Biology
Harvard University

The cellular evolution underlying variation in innate social behavior across Peromyscus deer mice

Understanding how genes can ultimately code for social behavior is an open question in biology. Variation in innate social behaviors across species offers an opportunity to use a comparative approach to identify the genetic building blocks of behavioral traits. Here, I investigate the variation of a sexually dimorphic social behavior – parental care – across Peromyscus wild mice. Parental care is carried out by only female animals in promiscuous species, but is shared across the sexes in monogamous species of Peromyscus mice. Using single-cell gene expression sequencing, I identify how differences in the cellular and molecular composition of the hypothalamus contribute to vast differences in social behavior of two Peromyscus species. I identify multiple cell types with abundance differences across species, including two cell types previously implicated in parental care. Furthermore, I find that the monogamous species has significantly fewer sexually dimorphic genes compared to promiscuous species, suggesting that changes in sex-biased expression may underlie evolution of sexually dimorphic behaviors. Finally, I find the changes across species are highly enriched to involve neuropeptidergic genes. Together, my data suggest the evolution of social behaviors involve changes in both cell abundance and sex-biased expression, with neuropeptide regulation likely playing a crucial role.