‘Insights to biological problems need multiple arsenals’

Dr Partha Majumder, the founder director of the National Institute of Biomedical Genomics has recently been elected to the Membership of the International Statistical Institute, The Hague, in recognition for his contributions to statistics. He spoke to Chief Editor of DBT Communication cell on his plans to accelerate interaction between various disciplines of science for better understanding of human health and disease.

You have been the Director of NIBMG since its inception. Now you have become an elected member of International Statistical Institute. Can this lead to collaborative research between these two institutes.

The International Statistical Institute (ISI; not to be confused with the Indian Statistical Institute, Kolkata), The Hague, is essentially an umbrella organization. The Institute promotes statistical science, formulates statistical policies and facilitates networking among statisticians and statistical organizations. The one plan that I have is to get into a long-term understanding with the ISI for organizing training programs in statistical genetics at the NIBMG. Initiation of joint training activities will also place members of NIBMG’s faculty in touch with statisticians working in human genetics in other global institutions, which will surely serve as a catalyst for collaborative research engagement.

What are the areas in which the two institutes can gain or learn from each other?

NIBMG is engaged in generation and statistical analysis of massively parallel DNA and RNA sequence data. These activities are throwing up massive data sets and complex statistical problems. Development and fostering of relationships between ISI and NIBMG will help find solutions to the statistical challenges of analysing and drawing inferences from massive inter-related genomic and proteomic data sets relevant to human health and disease.

Boundaries of disciplines are merging with interdisciplinary research becoming the way forward. What does our country need to do to integrate this spirit of interdisciplinary research into our system?

Like “doctors without borders,” NIBMG is an institution without Units or Departments. We have done so deliberately, in recognition of interdependence of various disciplines of science in our quest for understanding human health and disease. What does our country need to do? Here is a brief recipe:
(a) Encourage collaborations among scientists of institutions with disparate scientific foci by providing institutional support, such as adjunct professorships, fellowships to work across institutions, formal access to technology platforms across institutions, etc.;
(b) Organize inter-disciplinary training programs for graduate students even if short-term, such as summer or winter schools; and
(c) Provide larger grants for inter-disciplinary research.
Having said this, it is my belief that formation of interdisciplinary teams of scientists will be accelerated by the increased understanding that the complexity of biological problems requires multiple types of arsenal for gaining insights.

Are there robust examples of interdisciplinary research that the country could learn from? How would you set examples that can be emulated later?

Statisticians and biologists have always collaborated to solve problems in human genetics. This example is very robust and long-standing. Long-standing, because it is a natural coalition. Because of ethical and other issues, you cannot design experiments with humans. Necessarily, therefore, inferences have to be made from observational studies. Inferences from such data cannot be drawn without using statistical methods. I really don’t think that any examples of interdisciplinary research need to be set for later emulation. What is most important is to encourage deeper biological investigations, which necessarily will have to use concepts and methods from multiple disciplines of science. “Go deep and fund well.”

What are the still unaddressed research challenges in your area—population genetics in general and genetics of cancer in particular in the Indian context. Suggest a few steps that you will take to address them and how will your new position help that.

The greatest research challenge in biomedical genomics is to find methods for extracting high-quality and robust information from massive data sets. I don’t have a roadmap to do this! Concerted efforts by statisticians, biologists, bio-informaticians, computer scientists, etc. will be required. I am hoping that I will be able to negotiate with the governors of ISI to invite eminent statisticians and statistical geneticists to NIBMG and have them engage with our scientists. I also plan to facilitate eminent statisticians visit other biology institutions in India and engage with their scientists. Through engagement, there will be scientific benefit.

What are the human capacity challenges and the technological challenges and how would you like to address them. Will the new membership help?

Quantitative thinking, statistical science and computation have become integral to biomedical genomics, even of biology in general. Students in India pursue higher education in biology if they are not comfortable with mathematics. Unfortunately, when they get into biological research, they find themselves back into numbers, formulas and equations that they had dreaded in high school! Thus, there is a human capacity challenge. Again, I do not have a roadmap, but hopefully through coursework and short-term training during their biological research pursuits, they can be made to learn and improve their quantitative skills. Interestingly, increasing numbers of mathematicians and statisticians are taking interest in biological research, which is great. Technology manufacturing is in the hands of a small number of companies, whether we like it or not. All institutions in the world purchase their technology platforms from these companies. The challenge for us is to have enough money to purchase these platforms and reagents! Timely access for acquisition of these platforms is no longer a problem.

With your new esteemed membership how will you bring about an improvement in funding to make your research and institutional dreams come true

No ideas here!

What is your promise and hope for young statisticians to encourage them take up new fields of application and researchers from other disciplines to encourage them to develop statistical capabilities that can help them in their research? Can ISI play a constructive role in supporting both these groups?

It is really heartening that young statisticians are increasingly getting into applied research in biomedical sciences. They are seeing opportunities in massive data sets that require innovation in analysis and integration. ISI has been playing a major role in setting directions about the roles of statisticians in “big data science.” ISI and the Royal Statistical Society had organized a discussion meeting in London last year on “big data” (I was privileged to be invited to the meeting and to share my experience in “big data” analysis), which has resulted in a white paper charting the future course of statistical science. Hopefully, NIBMG will engage with ISI more thickly to accelerate interaction of biomedical genomics and statistical science.