Setting up an independent research group is not easy. Setting up interdisciplinary ones that are amalgams of group members coming from diverse academic backgrounds and speaking different disciplinary languages is even more difficult. However, this diversity, if channeled well, can drive cross-pollination of ideas and lead to valuable insights that none of us could have arrived at independently. Here, I share my journey of setting up a research group with members from backgrounds in biotechnology, mathematics, engineering (mechanical, electronics), bioinformatics, and physics, all focused on different aspects of gaining a better understanding of the dynamics of cancer metastasis and drug resistance – the two major clinically-unsolved challenges in cancer.

My first stint with interdisciplinary research was at the Indian Institute of Technology Kanpur, where I came across many faculty members from physics, mathematics, and engineering departments conducting research on biological systems. Motivated by them, I took up the challenge of working on mathematical modeling of cell polarity for my Master’s thesis under the guidance of Professor Pradip Sinha. While he headed an experimental biology laboratory, he gave me complete academic and intellectual freedom to identify a research question to be addressed by mathematical modeling, freedom that was of paramount importance. Reading seminal research papers that applied mathematical models to answer fundamental questions of developmental biology piqued my interest in systems biology and motivated me to pursue it for my Ph.D.

My colleagues in graduate school at Rice University came from previous training in physics, chemistry, and engineering disciplines. Together, we developed mechanism-based mathematical models to better understand how cancer cells adapt (switch their ​‘states’) during metastasis. The model predictions yielded useful insights; however, I felt incomplete by just making those predictions, without them being tested directly on cancer cells. Thus, in discussion with my advisors (renowned theoretical biophysicists Professor Herbert Levine and (late) Professor Eshel Ben-Jacob), I established and led collaborations with multiple experimental cancer biology groups to test the predictions. That enriching collaborative experience left an indelible mark on me in terms of how I would prefer to steer my independent research career. 

In October 2018, I moved to the Centre for BioSystems Science and Engineering at the Indian Institute of Science to start my independent research group. We started our journal club meetings in mid/​late 2019 and found ourselves discussing research articles covering a wide spectrum of topics – from cancer genetics (lineage tracing in vivo) to mathematical modeling (stochastic modeling of state transitions). I noticed that during these meetings, some of the group members, who came from backgrounds different than what was being discussed, seemed ​“disconnected” occasionally, especially as we delved into the fine-grained technicalities of certain tools and techniques. By the time we could figure out how to engage better, COVID-19 related disruptions had hit us. Moving these discussions to an online mode abolished the spontaneity of discussions that used to happen in the lab, thus further curtailing the chances of getting to know each other and our skill sets better. Therefore, while most group members were making good progress in their individual projects (which were largely based on their previous training), the ​‘emergent property’ behaviour (the whole is greater than the sum of its parts) was largely missing. This lacuna bothered me.

As a first step, I had detailed one-to-one conversations with my group members and realized that while most of them were eager to work synergistically with each other, they often found it difficult to find a common ground to connect and collaborate. That is when we tried a new approach – instead of attempting to create a common ground within ourselves right away, we started looking outwards and engaging actively with other research groups with similar interests but using a complementary approach to ours. Many initial meetings with these potential collaborators were just brainstorming sessions, where we made a wish list of questions each of us would like to see getting addressed. Next, we discussed what each of us could potentially contribute to; then, the dots started connecting. In many such meetings, my group members — instead of me — led the discussion on the latest simulations they had done and received pertinent feedback from our collaborators. Eventually, over a year, we started having preprints together (see here and here) and concomitantly, there was an increasing cohesiveness in our group too. For instance, when one of the group members shared what he/​she learnt from his/​her collaborative experience, many others chimed in with similar epiphanies. Thus, these ​‘outside’ collaborations gave many of us a first chance to connect, and the ice began to melt away. 

Another catalyzing aspect for us has been the influx of undergraduate students who are more ​‘undifferentiated’ and do not have strong self-constructed disciplinary silos. We noticed self-organization in terms of groups of students from different backgrounds who worked together, bringing in their complementary perspectives. For instance, two groups of two students each (a biology major and a physics major) had research papers as co-first authors (see here and here). Some group members started to cross the disciplinary boundaries by themselves and led the entire project. Of course, having gifted graduate students and postdoctoral fellows who enjoy mentoring undergraduate students as/​when required has been a key strength as well.

Once, in discussion with lab members, as I was working on a talk, an analogy serendipitously struck me: isn’t working in an interdisciplinary manner similar to behaving like cancer cells themselves? Cancer cells are notorious shapeshifters; many of them can change their behaviour or ​‘identity’ reversibly to adapt and evade many challenges (drug treatment, immune system). Importantly, cells with different ​‘identities’ improve the ​‘fitness’ of the population as a whole by heralding co-operation and collaboration among them. Thus, if cancer cells use all the mechanisms at their disposal for their benefit, why should we, the cancer research community as a whole, restrict ourselves to a specific set of tools, ideas and perspectives limited to our previous training?

Put together, from my limited three-year experience of setting up an interdisciplinary research group—Cancer Systems Biology Laboratory—here are some of the lessons I have learnt:

• Do not hesitate to collaborate across fields

It is not uncommon to develop cold feet when entering a new collaboration. It takes time to develop and co-lead interdisciplinary collaborations, simply because people across disciplines speak very different languages (despite their best intentions). Imagine what would happen if two people who speak different languages (and have a limited vocabulary of another language common to them) were shut in a room and were asked to come up with suggestions on a problem in a short time!

• Do not be shy to walk the ​‘extra mile’ of communication

It is often easy to ​‘disconnect’ in an inter-disciplinary collaboration, as many individuals do not walk the ​‘extra mile’ to try to speak in a language that is more ​‘natural’ to their collaborators. Many successful collaborations have one or more polylingual individuals who can understand the strengths and challenges of diverse research fields, to help arrive at a common agenda. 

• Do not try to force-fit an identity to yourself or your work

While collaborating with theoretical physicists, I am a cancer biologist; while working with cancer biologists, I am a physicist. Such identity crises are the norm, rather than the exceptions, of interdisciplinary research. When some of my group members are often confused with how to describe their work, I tell them: ​“If calling yourself X instead of Y changes your research, please go ahead and call yourself X. If not, don’t be stuck in what to call yourself. Let your work define you and your identity. Don’t let your previous training-based identity restrict your work or your next brilliant idea.”