Carlo Maj from the University of Marburg will discuss using polygenic modeling to analyze the genetic architecture of multifactorial traits. He will present how these approaches can be used to predict the genetically driven components of complex phenotypes. The talk highlights the potential of these methods to bridge genomic research and genetic epidemiology using biobank data. Why it matters: Such methods could improve disease risk assessment and advance personalized risk management in the region if applied to local biobanks or datasets.
Eran Segal from Weizmann Institute of Science presented The Human Phenotype Project, a large-scale prospective cohort with over 10,000 participants. The project aims to identify novel molecular markers and develop prediction models for disease onset using deep profiling. The profiling includes medical history, lifestyle, blood tests, and molecular profiling of the transcriptome, genetics, microbiome, metabolome and immune system. Why it matters: Such projects demonstrate the growing focus on personalized medicine in the region, utilizing advanced AI and machine learning techniques for disease prevention and treatment.
Professor Eran Segal presented The Human Phenotype Project, a longitudinal cohort study with over 10,000 participants. The project aims to identify molecular markers and develop prediction models for disease using deep profiling techniques including medical history, lifestyle, blood tests, and microbiome analysis. The study provides insights into drivers of obesity, diabetes, and heart disease, identifying novel markers at the microbiome, metabolite, and immune system level. Why it matters: Such large-scale phenotyping initiatives could inform personalized medicine approaches relevant to the Middle East's specific health challenges.
The Human Phenotype Project (HPP), led by researchers from Mohamed bin Zayed University of Artificial Intelligence (MBZUAI), published findings in Nature Medicine detailing an understanding of the health-disease continuum. The HPP involves deep and longitudinal profiling of approximately 28,000 participants, collecting diverse data including medical history, lifestyle, blood tests, and molecular profiling. The project aims to create AI-based predictive models for disease onset and progression, and digital twins to simulate interventions. Why it matters: This research can transform precision medicine and preventative care in the UAE by creating personalized digital twins that can simulate interventions and predict health trajectories.
KAUST's Environmental Epigenetics Program (KEEP), led by Prof. Valerio Orlando, focuses on understanding how cells acquire and maintain memory, particularly in response to environmental factors. The research investigates the role of non-coding RNA and chromosomal components in regulating gene expression beyond the DNA sequence. Epigenetics explains how the same genome can be interpreted differently, allowing cells and organs to adapt to changing conditions. Why it matters: This research could provide insights into how environmental factors impact gene expression and cell function, potentially leading to advances in understanding and treating diseases.
Søren Brunak presented deep learning approaches for analyzing disease trajectories using data from 7-10 million patients in Denmark and the USA. The models predict future outcomes like mortality and specific diagnoses, such as pancreatic cancer, using 15-40 years of patient data. Disease trajectories and explainable AI can generate hypotheses for molecular-level investigations into causal aspects of disease progression. Why it matters: This research demonstrates the potential of large-scale patient data and AI to improve disease prediction and generate hypotheses for further investigation into disease mechanisms relevant to regional healthcare systems.
MBZUAI researchers have developed MorphDiff, a diffusion model that predicts cell morphology from gene expression data. MorphDiff uses the transcriptome to generate realistic post-perturbation images, either from scratch or by transforming a control image. The model combines a Morphology Variational Autoencoder (MVAE) with a Latent Diffusion Model, enabling both gene-to-image generation and image-to-image transformation. Why it matters: This could significantly accelerate drug discovery and biological research by allowing scientists to preview cellular changes before conducting experiments.
Weizmann Institute Professor Eran Segal presented his work on the Human Phenotype Project at MBZUAI. The project is a large-scale biobank with data from over 10,000 participants, integrating medical history, lifestyle, and molecular profiling. Segal aims to use this data to develop personalized disease prevention and treatment plans. Why it matters: This research highlights the potential of interdisciplinary collaboration and big data analysis to advance personalized medicine in the region.