The picture above is an artistic rendering of a neuron (released under an open license). The rationale behind choosing this image for our "banner" figure was manifold. It is representative of our biological studies, particularly in the area of quantitative systems biology. It is also representative of the inspiration for artificial neural networks and machine learning in general, areas of research we are very active in. The capabilities of the neuron and what it has evolved to do is also indicative of our interest in evolutionary dynamics. More fundamentally, it represents the abilty of humanity to think rationally and solve problems.

About Us

What We Work On

Welcome to the Srivastava Lab website. The focus of our research is how to model and engineer molecular systems, including biomolecular ones. In fact, our group has traditionally worked in the area of systems biology and metabolic engineering. Recently, however, we have been using the computational tools we have developed to make forays in to the world of material science, specifically computational material design.

The reason we are able to work in two seemingly disparate areas is because of the type of computational frameworks and algorithms we develop and use. At the end of the day, we are interested in manipulating and engineering systems at a molecular level, whether or not they are biological. Our tools, which include mechanistic modeling, machine learning, artificial intelligence, and evolutionary algorithms, are applicable to a wide variety of areas, including molecular-level modeling and engineering.

What Drives Us

The fundamental idea that separates us from many other groups and drives many of our questions is our interest in evolutionary dynamics. It is the thread that unifies the various projects that we work on. In many respects, evolution is about optimizing system performance under a given set of constraints. Essentially, this is what an engineer is also trying to accomplish. The question we are curious about is if we can learn and/or harness evolutionary dynamics to allow us to engineer systems, whether natural or man-made. The field of evolutionary computation and its sub-discipline, evolutionary algorithms, seeks to explore this idea. Some of the better known evolutionary approaches include genetic algorithms and genetic programming. Other nature-inspired approaches often linked to evolutionary approaches and of interest to us include areas such as particle swarm optimization and ant colony optimization among others.

Additional Details

More detailed information about the type of work we do in the areas mentioned above may be found on our research page.

News

6/22/2020 - We've been fortunate to have been getting a lot of press about our work on COVID-19. There were two nice articles published in UConn Today here and here. Prof. Srivastava was also interviewed by a local NBC news affiliate.

4/14/2020 - Happy to be able to contribute to the effort in dealing with the COVID-19 pandemic. A nice article about our work can be found here.

3/04/2020 - Our recent paper on using machine learning to evaluate how pressure effects methane adsorption by MOFs was selected to be part of the The Journal of Physical Chemistry virtual special issue "Machine Learning in Physical Chemistry."