CMIT | Tangent Tuesdays 20th August 2024

Speaker: Suhana Nujum ( I2 B20, interned at University of Pittsburgh)
Title: Leveraging Single-Cell RNA Sequencing and Machine Learning to explore MicroRNA Regulation in Kidney Transplant Rejection
Date and Time: 20th August 2024 | 9 pm
Venue: PSB 1207

Abstract:Role of miRNA regulation in Antibody-Mediated Kidney Transplant Rejection: Antibody-mediated rejection (AbMR) is a leading cause of kidney transplant failure, with donor-specific antibodies (DSA) playing a critical role. However, only 40% of patients having DSA show rejection (DSA+ AbMR), while the remaining 60% are non-rejectors (DSA+ NR). MicroRNAs (miRNAs) are small non-coding RNA molecules that regulate gene expression. Our study aimed to identify key miRNAs and their target genes that influence rejection outcomes. We utilized a novel interpretable ML method developed in our lab, SLIDE (Significant Latent Factor Interaction Discovery and Exploration), to identify key miRNAs associated with rejection. Analyzing single-cell RNA sequencing (scRNA-seq) data from patient samples, we applied data analysis techniques and interpretable machine learning to identify significant genes linked to rejection. By identifying miRNA-targeted genes and understanding their involvement in the immune response, this study seeks to uncover potential biomarkers and therapeutic targets for improving graft survival.

Speaker: Sinta Maria Siby(B20, interned at University of Würzburg)
Title: Optimising tissue network configurations
Date and Time: 20th August 2024 | 9:30 pm
Venue: PSB 1207

Abstract: Over the years many mechanistic models have been developed to explain biological phenomena. However, the use of such models for practical purposes is limited since it requires adjusting the model's free parameters to be consistent with experimental observations. Several parameter estimation techniques are available to improve the quality of inference from models. My project explores the usefulness of a PyTorch-based package called sbi, which uses simulation-based inference algorithms based on neural networks for estimating the parameters of a 2-D tissue growth model (Simon Schardt, 2023).