BiMoDe - Modeling and Design for Biology
Human body is characterized by many parameters such as pulse, blood glucose level and lipid profile. At the scale of individual cells, there are no such established parameters. We are developing parameters and techniques for characterization of single cells. Since at the scale of individual cells, biochemical measurements such as protein expression and gene expression are tedious and expensive, we resort to parameters from physical properties. We focus on developing parameters from the shape and mechanical properties of cells and their organelles. For deriving parameters from shape, we use mechanical modeling and for estimating mechanical properties, we design micro-robotics platforms based on compliant micromechanisms. We are actively looking for collaborators and students to work on our objectives.
Micro-robotics platforms for manipulating single cells
We have designed and fabricated two micromechanisms for uniaxial and biaxial stretching of single cells. We converted the cell manipulation problem, uniaxial and biaxial stretching, into a mechanism design problem and designed an appropriate compliant mechanism for that actuation mode. In contrast to existing techniques such as Atomic Force Microscopy, Optical Stretching and Micropipette aspiration, which are limited to a single actuation mode, our method has the potential to combine multiple actuation modes on a single chip. By further developing other modes of cell manipulation, such as twisting and shearing, and incorporating them alongside the uniaxial and biaxial stretchers, we aim to create a lab-on-a-chip platform for comprehensive mechanical testing of single cells. Combining this mechanisms platform with our mechanics-based computational methods could pave the way for multi-modal physical characterisation of single cells to enable mechanics-based disease diagnosis.
Nondimensional parameters for characterizing nucleus shape
We assumed a simplified mechanical model for the nucleus as an inflated spherical membrane compressed between two rigid flat plates. By simulating the nucleus envelope using mechanics of membranes, we derived two nondimensional parameters - scale factor and flatness index for characterising nucleus morphology. Our model predicts a relationship among projected area, surface area and volume of individual nuclei, which was verified on cancer cell lines from the breast, liver and cervix. In other words, given the projected area and surface area of any individual nucleus from these cell lines, our model could predict its volume with less than 5% error [Balakrishnan et al. 2019]. We further performed a shape-variability analysis among a thousand nucleus shapes from these cancer cell lines. The highest variability modes were scaling and flattening (64% and 21% variability respectively), which highly correlated (> 0.9) with the scale factor and flatness index respectively [Devulapally et al. 2022]. Hence, these independent methods, mechanical modelling and variability analysis, are converging to same set of nucleus shape parameters. Through multiple experimental studies on multiple cancer cell lines, we show that the flatness index correlates with actin tension [Balakrishnan et al. 2021] and scale factor inversely correlates with the nucleus stiffness [Balakrishnan et al. 2019]. Actin tension and nucleus stiffness are known to indicate metastasis potential and ability to migrate through small pores, which are important considerations for cancer prognosis. Our results suggest these cellular properties can be conveniently obtained from nucleus shape by calculating our parameters - flatness index and scale factor. We are currently working towards developing such parameters for cell shape.
