In The Lab: Asking the right questions

-A A +A


In 1980s Kharagpur, India, the patterns dotted on the punch cards caught Tanmoy Bhattacharya’s eye.

Scattered across the paper, the clusters of tiny holes represented answers. The computer had all the knowledge. He just had to ask the right questions with the cards to get it — “talk nicely to the machine,” as he likes to say today.

Working in a laboratory, Bhattacharya began to see the broader picture of how an IBM 1620 — a very early computer that was loud, large and cumbersome with flashing lights and dials — thought and worked.

Head down, he clicked away, perforating his cards on a typewriter-like machine. All the little notches would create programming and data sequences for the computer.

This work was tedious, but it helped him understand the machine’s logic and how it functioned, from the hardware to the operating system. 

And it ultimately propelled his career toward the pursuit of some of the most advanced science ever known.

“Once you know the difficulties of designing a language, the rules of the computer language are no longer arbitrary. Rather, they become ingrained,” Bhattacharya said.

Over time, he found more opportunities to work on microprocessors and multi-terminal mainframe computers — not easy in a developing country. He mastered them all.

From punch cards to seeking solutions to the HIV crisis

Decades later, Bhattacharya, of Los Alamos’ Nuclear and Particle Physics, Astrophysics and Cosmology group, now helps maintain and program computers at Los Alamos that crunch complex data to design computational models for vaccines that can predict and possibly prevent HIV — one of the most enduring, complicated and devastating bloodborne diseases on the planet.

This work contributes to the development of the world’s first computationally generated vaccine sequences — including the HIV “mosaic” vaccine that was developed by Bette Korber’s team at the Lab and is one of only a handful of potential preventive measures for the disease to ever reach human efficacy trials.

Creating and administering programming data for vaccine sequences is a far cry from shuffling through decks of punch cards in Kharagpur. But the tenets of the old analog systems — that “ingrained” language Bhattacharya absorbed early in his career — still deliver critical knowledge.  

“My cordial relationships with computers stem from this complete picture, so often I can intuit why the computer did not give me the ‘logically’ correct answer,” he said. “I just place myself in the shoes of the designer/compiler writer/language specifier/etc. and ask myself to debug what I might have messed up in that role.”

Being able to step back and see the big picture and then taking details out of it to solve the larger problem has helped drive Bhattacharya’s career. 

Explaining the impossible

Physics is Bhattacharya’s favorite subject of all.

In 2000, he and colleagues started to examine the emergence of classical physics from quantum mechanics, asking why everyday objects seem to have concrete attributes like position.

Using the supercomputers of the day, their work showed that when measurements are neither very strong nor weak, there is no way to tell the difference between the classical and quantum models.

“In other words, under ordinary conditions, the world looks both classical and quantum, but since the classical description needs less mental resources to work with, we have learned to think of it as classical,” Bhattacharya said. “This body of work naturally led to the really interesting case when measurement is basically absent, so that quantum mechanics in its pure form flourishes.”

Not surprisingly, computation runs through many of Bhattacharya’s physics pursuits. 

Today, he is calculating the low-energy properties of neutrons from the current, reigning Standard Model of particle physics, which cannot account for things like dark matter or that the universe has matter in it, not just light.

“The standard model, for all its remarkable successes, cannot be complete: it does not explain the observed dark energy or dark matter, but even more importantly, it declares our universe, that we know exists, an impossibility,” he said.

“According to the Standard Model, matter cannot be present in the amounts necessary to create anything we see. We need hints at where the theory needs to be modified. Studying neutron properties and comparing them to precise calculations is the way to go,” he said.

Bhattacharya and others hope large scale computation can bridge these yawning gaps. 

“These are difficult calculations and they really need large-scale computation to make progress,” Bhattacharya said. 

Such diversified experience in both quantum mechanics and calculations in high-energy physics recently landed Bhattacharya’s team a grant from the Department of Energy to develop methods for using quantum computers to do high-energy physics.

Tanmoy Bhattacharya is in the Laboratory’s Nuclear and Particle Physics, Astrophysics and Cosmology group.