Nilay Kumar Email & Phone Number

Notre Dame, In, Us

• Led a team of 8 people in a cross collaborative project to develop and calibrate a novel, realistic, and multiscale mechano-chemical models of epithelial morphogenesis to infer general design principles that operate across multiple length and time scales (Manuscript in revision at Nature Communications) • Worked towards development and calibration of a subcellular element model for decoupling several factors of cytoskeletal regulation in generation and maintenance of organ shape (Manuscript published at PLOS Computational Biology)• Worked within a mixed team of applied mathematicians and biologists to develop an analytical partial differential equation-based network representation of interactions between morphogens, Rho family of GTPases and regulators of tissue mechanics. (Manuscript in preparation).• Participated in cross-functional teams to develop a cloud computing project for managing 13,000+ image datasets, applying bioinformatics, data processing, and data science methods to identify 4 lead therapeutic candidates using Python and MATLAB (Manuscript under review; arxiv preprint available)• Created a vertex model of epithelial morphogenesis, simulating the biophysical processes regulated by Piezo channels. The model was used to validate the experimental data predicting mechanosensitive Ca2+ channel Piezo's contribution in regulating robustness in organ growth. Developed numerous tools using MATLAB for automated quantitative analysis of large 3D fixed tissue and live imaging datasets. (Manuscript in preparation)

Notre Dame, In, Us

• Developed an automated framework using Bayesian optimization and Gaussian process regression for determining optimal physical interactions to create desired organ shape, resulting in a $48,000 pilot grant from Northwestern University's NSF-Simons Center for Quantitative Biology. Manuscript in preparation. • Created a machine learning based tool MAPPER (Github Link) that automates the process of analysis of adult Drosophila wing. Transfer learning was employed to retrain a Convolutional Neural Network (CNN) for the task of segmentation of insect wings. Classification algorithms were applied to classify and extract meaningful geometrical features from the regions of interest within the wing blade. Published in Frontiers in Genetics.• Reviewed advances in AI and data-driven statistical learning algorithms developed for quantitative analysis of multicellular system. The work resulted in a review article published in Journal of Biological Engineering.

Notre Dame, In, Us

• Contributed to the development of a low-cost microfluidic platform, capable of studying problems related to mechanobiology, reducing initial investment cost by 98.5% (from $20000 to $300) (Manuscript published in Biomicrofluidics)• Developed a novel simulation to study Taylor flow in micro channels, providing insight into dynamics of heat and mass transfer properties. Published in two prestigious journals, International Journal of Heat & Mass Transfer and Asia Pacific Journal of Chemical Engineering.

• Developed a fully trained automated pipeline to track real time mice activity and determine the motion patterns using a deep learning-based tool DeepLabCut• Involved in processing of neural recordings to establish correlations between stimulation of vagus nerve and movements of mouse

Delhi, New Delhi, In

• Worked towards designing a lightweight liquid armor in a Govt. of India sponsored project worth INR 20 million (USD $240,000) funded by the DRDO.• Developed a code in FORTRAN for the project and ideation, fabrication, and installation of experimental setup for results validation . • Used Stokesian Dynamics to study the effect of far field hydrodynamic interactions and frictional coefficient in shear thickening. simulate a system consisting of a mixture of rods and spherical particles and studies its impact on shear thickening.• Worked on device fabrication for flow visualization experiments to observe the microstructure of suspensions exhibiting shear thickening behavior.

Bochum, Nordrhein-Westfalen, De

• Designed the geometry and generated an unstructured mesh using FreeCAD and Gmsh for the studyof the flow through cerebral aneurysms.• Integrated the above generated mesh in complex C++ code, which combines the Lattice Boltzmanmethod for fluid flow with the finite element method and immersed boundary method for the dynamicsof flexible membranes.

Doctor Of Philosophy - Phd, Chemical And Biomolecular Engineering

Bachelor Of Technology - Btech, Chemical Engineering