Car park to lab in Covid: Infosys co-founder's bet put IIT Madras on brain map

The Sudha Gopalakrishnan Brain Centre is an IIT Madras laboratory built to create three-dimensional maps of human brains at cell-level detail. It matters because ordinary brain scans cannot show individual cells, limiting knowledge of neurological disease. The centre aims to examine brains across development and ageing.

Its origins trace to Kris Gopalakrishnan, an Infosys co-founder and IIT Madras distinguished alumnus, who asked whether the institute could attempt something ambitious. The project was approved in 2018, with investment from Gopalakrishnan and support from the Principal Scientific Adviser's office. The team adapted a former car park into laboratory space.

The money arrived in February 2020, but the campus closed the next month. Engineers and neuroscientists continued working through the disruption, and everyone on the team caught Covid-19, according to Professor Mohanasankar Sivaprakasam. The centre was inaugurated in 2022, with its machines running.

Mapping a human brain cell by cell means recording the locations and structures of cells throughout the entire organ, rather than viewing only its broad shape. The centre creates three-dimensional maps at micron-level resolution. This matters because different diseases may affect particular cell types, and researchers need to see where those cells are and how they are arranged.

The process begins by cutting one brain into about 10,000 sections. Some sections measure as much as 15 centimetres by 20. Each piece is photographed at very fine resolution. A computer then puts the images back together, recreating the brain as a digital three-dimensional volume.

This is not simply a larger scan. The resulting files can run into petabytes, or about a million gigabytes. The centre had already captured whole-brain, cell-resolution volumes of three developing brains by its inauguration. Its mission is to map more than 100 human brains.

The Sudha Gopalakrishnan Brain Centre's mission is to build three-dimensional, cell-resolution maps of more than 100 human brains. The planned collection is intended to cover brains from the earliest weeks of life to old age. Such a range could reveal how brain structure changes across development and ageing.

The scale of each map is enormous. A brain must be divided into about 10,000 sections, some as large as 15 centimetres by 20. Each section is photographed, and the images are computationally joined into one whole-brain volume. This creates a demanding storage and processing task.

The article says the data from one brain runs into petabytes. One petabyte is about a million gigabytes. By the centre's inauguration, the team had captured three developing brains at whole-brain, cell resolution. The centre also needs the process to be fast; Professor Mohanasankar says one brain a year would not be enough.

MRI scanners use magnetic fields to picture the inside of the body, but their resolution is relatively coarse for studying individual brain cells. The article places MRI resolution at about one millimetre. A neuron measures around 10 microns, and a micron is one-thousandth of a millimetre. That difference prevents an MRI from showing single cells.

The brain centre uses a much finer imaging approach. It cuts whole brains into thousands of sections and photographs each section at cell resolution. Computers then reassemble the images into a three-dimensional volume. The result can show cells across the brain, rather than only larger regions or general patterns.

This finer view matters because the brain contains 80 to 100 billion neurons and many more supporting cells in roughly 1.3 litres. To understand or treat a disease, Professor Mohanasankar says researchers need to know which cell type is affected and how. The centre's resolution is about a thousand times finer than MRI's.

Kris Gopalakrishnan began asking whether IIT Madras could undertake a major brain project. The Infosys co-founder and distinguished alumnus provided the seed and vision, then took the first investment risk. The project grew from an earlier IIT Madras effort to image human brains outside the body at terapixel resolution.

IIT Madras leadership turned the question into a formal project. Then director Professor Bhaskar Ramamurthi agreed to proceed, and the decision was taken in 2018. Professor Mohanasankar Sivaprakasam led the engineering and neuroscience effort. The team built machinery that could handle human brains, rather than relying on systems designed for much smaller mouse brains.

Professor K VijayRaghavan, then the Government of India's Principal Scientific Adviser, came on board. His office also backed the project. Despite campus closure during Covid-19, the team converted a former car park into laboratory space. The centre was inaugurated in 2022, and its machines were operating.

The main promise is greater precision. Current brain imaging can show broad structures, but it cannot normally identify individual cells. Cell-resolution maps could reveal the kinds of cells involved in a disorder, their locations, and how their surrounding networks are organised. That information could improve the biological basis for studying neurological disease.

The article gives a simple mechanism: the brain contains many cell types, and a disease may affect one type more than another. The centre photographs thousands of thin sections and digitally rebuilds the whole brain. Researchers can then examine the affected cells within the complete three-dimensional context, instead of studying only a small area or a broad scan.

The article does not claim that these maps already cure disease. It says neurology and neurosurgery are held back by limited knowledge of the brain itself. Comparing detailed maps from more than 100 brains, spanning early life to old age, could provide a stronger foundation for future research and treatment.

Neurons are cells in the brain that form its primary nerve-cell population. The article also describes many supporting cells packed alongside them. Together, these cells fill about 1.3 litres of brain tissue. The centre studies both the enormous number of neurons and the wider cellular environment around them.

The key issue is arrangement. The brain contains 80 to 100 billion neurons, plus many more supporting cells. A cell-by-cell map can show where different cells sit within the whole organ. The centre creates that view by photographing roughly 10,000 sections from each brain and computationally reassembling them in three dimensions.

This matters for disease research because, as Professor Mohanasankar explains, treatment requires knowing which type of cell is affected and how. Broad scans cannot provide that detail. Comparing complete cellular maps across brains could therefore help researchers connect particular cell patterns with development, ageing, or neurological disorders, although the article does not describe a completed treatment.