We adapt engineering principle in the realm of biochemistry and reverse engineer molecular biology of the cells. This allows us to create cells with unique capabilities. Our research includes
1. Can we make bacteria artificially intelligent (AI)?
2. How do computation and plasticity arise in simplified neural microcircuits?
3. Creating smart therapeutics and novel cellular sensors with synthetic genetic circuits.
1. Can we make bacteria artificially intelligent (AI)?
We built living bacterial systems capable of abstract, algorithmic problem solving, including mathematical and linguistic tasks previously considered exclusive to silicon-based computers. In a recent study (Nature Chemical Biology, 2024), we built multicellular bacterial computers capable of identifying prime numbers, vowels, perfect power numbers, and connected Haar graphs, and of solving mathematical problems such as determining the maximum number of pieces obtained by straight cuts of a circle. The system also answered mathematical queries, including whether n! is divisible by n(n+1)/2 and whether n² can be expressed as the sum of three factorials. Bagh built bacterial biocomputers, that solved abstract maze problems (ACS Synthetic Biology, 2021). Such tasks were previously unachieved in engineered living systems.
To build such ‘intelligent’ bacteria, Bagh established artificial neural networks (ANN) with living bacteria as a new foundational platform with streamlined mathematical and biochemical design rules (Chemical Science, 2021). Individual bacteria function as artificial neuro-synapse that collectively form functional ANNs in cultures, enabling implementation of computation. Problems were posed as binary combinations of chemicals and the answers were read through defined fluorescent protein expression patterns.
A double Feynman Gate with Cellular Artificial Neural Networks Wetware
Richard Feynman once suggested that biological cells could be a choice for reversible computing due to their very low energy budget compared to an electronic computer. We built an artificial neural network using genetically engineered E. coli, such that they could perform complex computations and form a double Feynman logic gate. Double Feynman gate is a form of reversible computing, which is different from conventional computing.There are five sets of genetically engineered bacteria that, in a test tube, work as a biocomputer and form artificial neural networks. The biocomputer takes three chemicals as input signals and produces two light-emitting proteins as output, such that bacteria start glowing. Thus, based on the chemical signal present in the environment, the population of bacteria would decide which protein needed to be expressed. These bacterial artificial neural networks decide and work as a double Feynman gate (Rajkamal Srivastava and Sangram Bagh* (2023) , A Logically Reversible Double Feynman Gate with Molecular Engineered Bacteria Arranged in an Artificial Neural Network-Type Architecture, ACS Synthetic Biology, 2023, 12, 1, 51–60
Can we build a biocomputer with engineered bacterial community to solve mathematical problems?
We synthetically engineered a set of bacteria, which work as a distributed biocomputer to solve a set of computational/mathematical maze problems. This is the first time we are showing engineered bacteria can solve abstract mathematical/computational problems. This work has significance in creating new technology for cellular cryptography and steganography (Sarkar, et al ACS Syn Bio, 2021). The work has been featured in MIT Technology Review. https://www.technologyreview.com/2021/11/09/1039107/e-coli-maze-solving-biocomputer/
We are developing a platform based on engineered neural progenitor cells to construct simplified neural microcircuits in vitro. This system will enable controlled investigation of how computation, memory, and adaptive dynamics emerge in living networks.
Rather than reproducing the full complexity of the brain, we focus on minimal architectures that allow quantitative analysis of network-level computation and plasticity. These tractable neural systems provide a foundation for exploring the biological principles underlying intelligent behavior.
Complex electronic analogous genetic device in single bacterial cell to process higher order information
We develop strategies and methods to create electronic analogous complex synthetic genetic circuits within single bacterial cell to perform higher order information processing. Our recent work on creating 3-input-3-output integrated circuits has been published as a front cover page article in Bioconjugate Chemistry, 2020 https://pubs.acs.org/doi/abs/10.1021/acs.bioconjchem.9b00517 (Front Cover Page Article)
3. Creating smart therapeutics and novel cellular sensors with synthetic genetic circuits.
Cancer gene therapy with bio-bots
We are developing a cellular robotics platform for cancer gene therapy. We reprogram microbial cells as 'robots' (biobots) to sense varieties of extracellular signals similar to a cancer microenvironment, to decode those signals, to co-ordinate with other cells and decide to invade cancer cells. We apply those 'bio-bots' for programmed delivery of therapeutic genes and RNAi into the cancer cells. See our recent publication (Srivastava R. et al, ACS Synthetic Biology, 2022, 11, 3, 1040–1048 https://pubs.acs.org/doi/abs/10.1021/acssynbio.1c00392.
This work is highlighted in Nature India (Engineered Bacteria Can Silence Genes in Cancer Cells) https://www.nature.com/articles/d44151-022-00032-6
Novel cellular sensors for Space Synthetic Biology
How about a set of synthetic genetic systems inside appropriate organisms, specially designed for microgravity conditions for potential uses in space stations and during exo-planet journeys?
Synthetic Biology has been proposed as a key component for regenerative and sustainable closed loop solution in long-term space habitation. Here we have developed the first microgravity biosensor using synthetic genetic systems inside appropriate organism. Our work is the first to integrate microgravity as a physical signal within cellular processes in a human designed way (Mukhopadhyay and Bagh, Biosensor and Bioelectronics, 2020). Our previous publication was featured in Nature India, May,2016 (http://www.natureasia.com/en/nindia/article/10.1038/nindia.2016.72) and Times of india.