Spring 2025 Student Seminars

March 19

Whitney Souery
Advisors: Dr. Marc Lenberg and Dr. Ehab Billatos
Title: Using the airway field of injury to investigate the pathophysiology of mucus plugging

Abstract:
Mucus plugging (MP) affects around half of people with COPD and is associated with significant airflow limitation and mortality. To improve our understanding of MP pathogenesis, we identified a gene expression signature of MP in bronchial epithelium to understand biological processes and clinical phenotypes associated with these gene expression patterns. We studied 204 participants from the Detection of Early Lung Cancer Among Military Personnel (DECAMP) 2 cohort. MP was scored as the number of lung segments with mucus plugs in study CT images. RNA sequencing of airway brushings of the mainstem bronchus from DECAMP 2 participants was previously reported. Genes differentially expressed with increasing MP score were identified by Spearman correlation. Gene clusters and participant subgroups were identified by clustering. The association between participant subgroups and available clinical annotations was explored to understand the phenotypic differences between MP-associated gene expression-defined participant subgroups. Our findings reveal differentially expressed genes in the bronchial epithelium of individuals with MP are enriched for epithelial and immune pathways that represent independent processes in MP with distinctive phenotypic associations. Our analysis revealed several MP gene clusters associated with chronic bronchitis and the co-presence of cough and phlegm, suggesting these genes may reflect pathways contributing to the development of bronchitic-related symptoms, or may represent consequences of the symptoms themselves.

 
Amulya Shastri
Research Advisors: Dr. Joshua Campbell, Dr. Stefano Monti, Dr. Joseph Mizgerd
Title: Sub-phenotypes of pneumonia: A story of host response, causal agents, mouse models and machine learning

Abstract:
Pneumonia involves heterogeneous immune and pathophysiological processes in the lung; however, host-response heterogeneity in pulmonary pathophysiology during pneumonia remains poorly characterized. We hypothesized that human pneumonias segregate based on host response. We used histopathology scores and immunohistochemistry signals to define pulmonary responses in autopsy samples from patients who died with or without diagnosis of pneumonia and applied ML algorithms to identify and characterize pneumonia sub-phenotypes. Mouse models are often used to study human pneumonias, yet we lack a direct comparison between the two. We scored same modalities in mice and applied the same ML framework to capture the relationship between human and mouse pneumonias. Our analyses indicate seven distinct pulmonary pathology sub-phenotypes of pneumonia.  Refinements to mouse models are needed to fully capture the diverse histopathological landscape of human pneumonias, but the mouse models examined will have promising utility for assessing the biology of pneumonia pulmonary pathology sub-phenotypes including underlying mechanisms and pathophysiological significance.

 
April 2

Hannah Lords
Advisor: Paola Sebastiani
Title: Genome Wide Association Analysis of 9 Neuropsychological Tests from the Long Life Family Study

Abstract:
Cognitive function reflects the interplay across memory, attention, language, and executive processes that enable information storage, retrieval, and application. While cognitive performance generally declines with age, a subset of individuals—healthy agers—exhibit delayed or minimal decline, suggesting an underlying genetic protection. In the Long Life Family Study (LLFS), we leverage this unique population to conduct a GWAS on neuropsychological test scores, capturing a stationary snapshot of cognitive performance independent of age-related deterioration. By analyzing each neuropsychological test separately rather than collapsing them into a composite measure, we preserve the ability to detect variants with specific domain effects, thereby offering a more nuanced and interpretable view of the genetic architecture underlying cognitive function. We performed genome-wide association studies on nine neuropsychological tests administered in the Long-Life Family Study, identifying 10 genome-wide significant SNPs (p<5e-8). Each SNP was unique to the test it was identified in. Furthermore, we leverage our test specific results to provide domain specific annotations to variants identified in a GWAS of general cognitive function performed in the UK Biobank.