Posters - Schedules

Posters Home

Parkinson’s Disease (PD) is a brain disorder that affects upwards of ten million people worldwide. It impairs motor and non-motor skills, leading to symptoms such as tremor and cognitive dysfunction. The cognitive symptoms of PD impair many higher-order functions associated with regions in the temporal and occipital lobes of the brain. While studies into the middle and later stages of PD have found abnormal activity in these regions, it is unclear how these findings translate to earlier stages of the disease. Current diagnosis methods of PD rely on observable physical characteristics, which means that detection is not as early as it could be. Neuroimaging biomarkers can help with detecting the disease earlier and more efficiently through non-invasive means. Here, we compared neuroimaging data from the Parkinson's Progression Markers Initiative database for PD patients and controls using graph theoretics and persistent homology analysis to understand how temporal and occipital brain networks in PD diverge from normal functioning at an early stage. Abnormalities across various metrics were found in temporal and occipital regions responsible for visual processing and other cognitive tasks known to be disrupted by PD. Divergence from normal functioning in these areas may indicate the development of certain Parkinson’s-related symptoms.

One of the common tools to explore interactions, associations and assemblies in microbial data are association networks. With today's state-of-the-art technology, they are often used within system-wide approaches to investigating relevant data that can function in multivariate, interconnected settings that account for important statistical features. However, most of these tools are still not without their issues, be it statistical and theoretical shortcomings like data compositionality and bias, computational bottlenecks in many existing tools or failing to account for important research contexts such as longitudinal experiments, complex interactions or heterogeneous data.

The following discussion aims to pinpoint the most important problems that current state-of-the-art tools struggle with, focus attention potential and existing solutions and consider new ideas. Additionally, tools that can supplement network analysis in different contexts, such as feature selection methods, simulation tools, data processing techniques should be kept in mind, as together with association network tools they produce a strong analysis pipeline. As such, approaches to exploring direct associations in heavily multivariate and heterogeneous data, that might have non-linear associations or problems such as zero inflation, compositionality, overdispersion, need to be investigated and a robust stepwise approach that would assist with the analysis of microbiome must be created.

The highly contagious SARS-CoV-2 and its associated disease (COVID-19) are a threat to global public health. To develop effective treatments for COVID-19, we must understand the host cell types, cell states and regulators associated with infection and pathogenesis such as dysregulated transcription factors (TFs) and surface proteins, including signaling receptors. To link cell surface proteins with TFs, we recently developed SPaRTAN (Single-cell Proteomic and RNA-based Transcription factor Activity Network) by integrating parallel single-cell proteomic and transcriptomic data based on Cellular Indexing of Transcriptomes and Epitopes by sequencing (CITE-seq) and gene cis-regulatory information. We apply SPaRTAN to CITE-seq datasets from patients with varying degrees of COVID-19 severity and healthy controls to identify the associations between surface proteins and TFs in host immune cells. Here, we present COVID-19db of Immune Cell States (https://covid19db.streamlit.app/), a web server containing cell surface protein expression, inferred TF activities, and their associations with major host immune cell types. The data include five high-quality COVID-19 CITE-seq datasets with a toolset for user-friendly data analysis and visualization. We provide interactive surface protein and TF visualizations across immune cell types for each dataset, allowing comparison between various patient severity groups for the discovery of potential therapeutic targets and diagnostic biomarkers.

Head and neck cancer is the most incident and prevalent cancer in men in South Asian countries, particularly in Pakistan. Prediction of pathological stages of cancer can play a critical role in early diagnosis and personalized medicine. This project ventures into predicting the different stages (Stage I-IV) of head and neck squamous cell carcinoma (HNSCC) by analyzing DNA methylation patterns. We prioritized 485,577 methylation CpG sites based on their highest predictive power using a wrapper-based feature selection method, along with different classification models. We identified 68 high-power methylation sites, corresponding to 67 genes, that predicted the pathological stage of HNSCC samples with 90.62% accuracy using a Random Forest classifier. Our results led us to unveil a dynamic stage-course network for HNSCC that evolves over the four cancer stages. This network displays potential candidates for each stage, with 37 persistent nodes and unique complexes, revealing their role in cancer-related pathways. We intend to study these candidates further to determine their putative role in HNSCC initiation and/or progression, using functional datasets from CRISPR, RNAi, and drug screens. This study offers a promising stage-course method to resolve cancer initiation and progression at a higher resolution to identify key proteins that have biomarker potential.

COVID-19 is a disease caused by the SARS-CoV-2 virus, where aging is a risk factor for severity. Methylated genes in blood have been proposed as markers for possible prognosis in older adults. This study explored the potential biological role of methylated genes as prognostic markers in COVID-19 for older adults. We analyzed 1535 peripheral blood samples from 10 experiments in the Gene Expression Omnibus database (88 controls and 1447 COVID-19). Experiments were individually analyzed using Bioconductor packages and the miRNet platform. We identified four possible biomarker genes: COL11A2, previously related to COVID-19 severity, C9orf125, which has been associated with other respiratory diseases, C17orf97, and GIT1, which, has no available information. We obtained ten regulatory networks involving differentially methylated genes, transcription factors, and microRNAs, considering other factors such as sex or aging only. Key transcription factors DNMT1, MECP2, and HDAC1 were associated with epigenetic marks, while various miRNAs were linked to respiratory diseases. This research provides insights into potential COVID-19 biomarker genes for older adults and their possible regulators, contributing to the understanding of the biological role of methylated genes in the disease.