Manolis Kellis on the Computational Biology of Human Disease, Genetic Perturbations, and Societal Interventions
Summary
Manolis Kellis, head of the MIT Computational Biology Group, discusses the profound complexity of understanding human disease, equating it to the intricate nature of the human genome, brain, epigenome, and immune system. He highlights a significant paradigm shift in biological research: while traditionally, basic biology was first understood through model organisms and then applied to humans, the vast amount of natural genetic variation in humans is now driving fundamental biological discoveries. The central challenge is to move beyond mere correlation to pinpointing causal disease mechanisms through the analysis of genetic perturbations.\n\nKellis explains that perturbations are the key to understanding any system. In animal models, researchers can perform controlled, strong, single-gene perturbations. In contrast, human genetics leverages the millions of natural genetic variants (each individual carrying approximately six million unique perturbations) that have accumulated through natural selection over millennia. He differentiates between traditional Mendelian genetics, which focuses on strong-effect, single-locus mutations within families, and modern approaches that analyze large cohorts of unrelated individuals, correlating millions of genetic variants with hundreds of phenotypes, often derived from complete electronic health records. He notes a crucial trade-off: detecting large, clear effects early in the biological pipeline (e.g., epigenomic alterations) means a longer causal path to the ultimate disease, whereas subtler effects closer to the disease phenotype (e.g., gene expression) have a shorter, more direct link.\n\nThe ultimate goal of understanding these disease mechanisms is to enable effective interventions, whether through targeted medications or lifestyle modifications. Kellis's group employs computational methods to deconvolve the enormous genotype-phenotype matrix, integrating diverse data from epigenomics, transcriptomics, and single-cell profiling across various tissues and cell types, such as different brain regions and neuronal types. He advocates for measuring a much richer array of phenotypes, spanning from molecular and cellular levels to organ-level imaging and comprehensive organism-level behavioral and cognitive tests, to meticulously trace the complete causal pathway to disease.\n\nKellis also touches upon the "importance" of diseases, considering both their impact on quality of life (e.g., mental disorders, Alzheimer's) and their mortality rates (e.g., heart disease, cancer, COVID-19). He emphasizes that every disease results from an interplay of both genetic and environmental components, and a deep understanding of the genetic mechanisms can significantly inform and enhance environmental interventions. He uses the analogy of a "bathtub of fat" for obesity to illustrate how internal metabolic processes (influenced by genetics) profoundly affect the external "in-out" equation of diet and exercise. He also briefly connects the profound complexity of biological systems to the challenges inherent in engineering artificial general intelligence.
Key Quotes
"irregularities standing human disease is the most complex challenge in modern science"
"human genetics has been so transformed in the last decade or two that human genetics is now actually driving the basic biology"
"perturbations is how you understand systems"
"every one of us carries six million perturbations"
"the world's greatest puzzle and you can now solve that puzzle by throwing in more and more knowledge about the function of different genomic regions"
"human genetics basically looks at the whole path from genetic variation all the way to disease"
"the most effective place where to detect the alteration that results in disease is earlier on in the pipeline as early as possible"
"every single disease has both a genetic component and environmental component"
"understanding the genetic component can have a huge impact even on the environmental component"
"your metabolism impacts that you know bathtub basically your metabolism controls the rate at which you're burning energy it controls away the rate at which you're storing energy"
Concepts
Themes
- Complexity of biological systems
- Paradigm shift in biological research
- Causality vs. correlation in disease understanding
- Multi-scale analysis of disease
- Interplay of genetics and environment in disease
- Societal impact and intervention in health
- The "puzzle" metaphor for scientific discovery
Related to:
Similar Episodes
Computational Biology of Coronaviruses: Understanding Viral Structure, Function, and Pandemic Preparedness
Jonathan Reisman on The Human Body: Evolutionary Design, Taboos, and the Doctor's Perspective on Anatomy
Manolis Kellis on the Digital Nature of Life, Human Genome's Beauty, and the Dynamics of Genetic and Horizontal Inheritance