Project 1: Shedding light on the genetic dark matter of the human microbiome
High throughput reverse genetics for microbiome gene function discovery
Hundreds of species of fungi, microeukaryotes and bacteria live in our gut as communities, and carry out beneficial functions for our health, including helping us to digest food, train our immune system, and defend against pathogens. While it is becoming increasingly clear that the human gut microbiota contributes manifold to human physiology, we have limited understanding of the underlying molecular mechanisms. A prime reason for this is the lack of relevant model species in the human gut microbiome. As a matter of fact, most genes encoded within the gut microbiome are of unknown function and have no homologues to current model species like E. coli. This lack of fundamental knowledge limits our ability to make functional predictions and rationally modulate the microbiome to design microbiome-based therapies. By joining forces, the Typas and Huang labs combine their expertise in bacterial genetics and high-throughput phenotyping to create systematic genetic tools that can be the basis for deorphanizing genes with unknown functions of important microbiome members. The ultimate aim is to deepen our understanding of fundamental bacterial functions carried within microbiome strains and to establish new model organisms for the human gut microbiome.
We are developing novel genetics tools and methodologies to generate high-quality and high-diversity genome-wide mutant libraries in members of one of the most impactful orders in the human gut microbiome; the Bacteroidales. The mutant libraries will be screened in high-throughput fitness and single-cell assays across hundreds of conditions in vitro as well as in animal hosts. Such systematic reverse genetics approaches generate a wealth of data from which we can extract gene functions and chart functional units.
Project 2: Sensing and responding to cell shape changes
The importance of size and shape:
Bacterial cells come in a wide variety of shapes and sizes. Understanding how bacteria maintain and change their physical form can have important applications for controlling bacterial growth, identifying targets for new drugs, and provide the blueprints for designing synthetic structures. Bacterial morphology plays a key role in many vital functions such as biofilm formation, motility, and pathogenesis, and bacteria tightly control shape across growth conditions and the cell cycle. Using bacteria as a model organism, we can gain a number of fundamental insights in basic biology.
In one line of research we are interested in elucidating how bacteria sense and respond to environmental changes that affect their cell size, and how they attempt to maintain homeostasis. Another focus is the function of the main cytoskeletal component defining cell growth and shape in rod bacteria, MreB. Here we are interested in how disturbing this essential structural component affects cellular physiology and the ability of cells to adapt to changing environments.
Project 3: Predicting Drug Action & Interactions Using Big Data
Understanding mode of action in drug discovery
In an era in which antibiotic resistance is increasing at an alarming pace and remaining active antibiotics become scarcer, we urgently need new effective compounds, or combinations that will improve the function of existing antibiotics.
Elucidating the mode of action (MoA) of small molecules targeting microbial growth is key for drug discovery. Knowing the pathways and proteins involved makes it possible to estimate drug specificity vs. side-effects, drug spectrum, and the risk of resistance development. MoA determination is still a major bottleneck in drug discovery because it depends on laborious low-throughput methods that cannot be applied to large compound collections. The complexity increases exponentially when investigating combinations of drugs and would require exhaustive screening. Together, the Typas and Huang labs seek to reduce the complexity in drug MoA studies by combining their expertise in high-thoughput systems biology (Typas), and investigations into how the physical structure of cells is established (Huang). By combining these approaches, they can direct their approach to interrogate specific aspects of bacterial physiology.
We are using high-content data (chemical genetics, quantitative microscopy) and structural data as input for computational network and machine learning analyses in order to identify sets of genes that can predict drug MoA and drug-drug interactions across a wide range of bacterial species. Our systematic datasets also yield novel insights into the pathways affected by drugs and how cells can counteract or bypass their action.