“In the long history of humankind (and animal kind, too) those who learned to collaborate and improvise most effectively have prevailed.”
― Charles Darwin
I started off as a basic science researcher in the 90s at the Indian Institute of Science, when research was an individual's initiative: a “solo” project. Even if we would develop or deliberate our ideas with fellow colleagues from other laboratories, which in my case did happen extensively, this dialogue was always a conversation, not a formal collaboration. Most laboratories were open to sharing resources or unpublished methods for technical issues. This help generally merited an acknowledgment but seldom authorship, in the ensuing publication. From this basic conditioning, I moved on to genomics sciences which in those days was fast evolving into a research enterprise that involved global collaborations across laboratories. Consequently, this culture became embedded in my own research career and a majority of my projects run in the “collaborative” mode. Having done all my research in India, spanning more than 25 years, in this essay I would like to share my perspective on the key ingredients for sustained collaborations.
Broadly, the impact of any research, whether for academic interest, societal benefit, commercial or translational, can be increased manifold via collaborations. However, there are trade-offs in solo vs. collaborative efforts. In solo mode, there is high personal satisfaction and sense of ownership. When the work is in the proof-of-concept stage or fundamental biology space, where the potential is not apparent or quantifiable, it is ideal to initially pursue it in solo mode. If the projects in “solo” mode however are not aligned with the institutional mandate, they are challenging to sustain.
The foremost advantage of collaborations is timing: the ability to go from idea to implementation, without losing out on the prime mover advantage. Since expertise, resources and infrastructures are shared, the projects also become more affordable. A drawback of collaborations carried out to build resources, databases or registries is the lack of immediate academic appeal and their perception as intellectually sterile exercises. Long-term collaborations also generate assets that are useful after the project tenure is complete. I illustrate these points with examples from my career.
I was part of the Indian Genome Variation Consortium which built a comprehensive resource of variations from diverse Indian populations. The first years of the project involved extensive sample collection and genotyping. This baseline catalogue unexpectedly became seminal in many later interesting discoveries, such as the linkage of geoclimatic adaptation (high altitude, salinity, humidity) with human phenotypes and disease, founder mutations and population histories, to name a few from our group. In another instance, we have been studying a group of neuro-degenerative disorders called ataxia with AIIMS New Delhi. Here, over the past 20 years we have built a genetic registry of more than 5000 patient families. A bulk of the time in the initial 10 years (a long time in a student or PIs career) went in genetic diagnosis of ataxia and creating visibility. In those times, it was a challenge to appreciate the potential for future research and equally challenging for researchers to participate in such projects. Fast forward 20 years: this clinical cohort has been crucial to investigate the genetic basis of rare diseases in India, which was not the initial goal of this project. Along with these projects, I still pursued one project in “solo” mode: understanding the role of Alu repeats in genome organisation and function, which has been deeply fulfilling.
How to collaborate?
First, the need for collaboration depends on the domain of the problem that needs to be addressed. For instance, most biological fundamental research can be done solo or with limited collaboration. If the needs are applied, then it might require an expertise of a different discipline for implementation as well as more people of the same discipline for scaling up. If the area of research is translational then one cannot move ahead without collaborations. For instance, our projects in the areas of ataxia disease genomics and Ayurgenomics would have been incomplete without the participation of modern and Ayurveda clinicians, data scientists and public health professionals. The kind of collaboration also depends upon the degree of innovation. For radical innovations there might be a greater need for functional diversity, than for incremental innovations.
A second aspect, which needs consideration during planning of collaborative projects, is the type of setting and infrastructure needs for collaboration. Studies which can be conducted in academic institutions or involve clinical settings/field-work have different requirements for infrastructure, resource and manpower.
A third important consideration is whether the collaboration is anticipated to be long or short term. It is very important to have an open discussion on resource sharing, terms of collaborations and IPR sharing policy at inception. Many times we think it’s too premature to deliberate on these issues but this is important for pre-empting future conflicts.
Fourth, a successful collaboration requires stakeholders, each with well-defined ownership as well as accountability. It is also important to respect the motivations for individual stakeholders. Drivers could be personal incentives such as: more acceptability by peers, awards or monetary benefits. It could also be driven by an organization’s need either for revenue or national as well as global recognition and visibility. Many times a need for societal relevance also could be a key driver. Stakeholders should not have conflict of interests and not get into a competitive position at each other’s expense. The diversity in expertise amongst stakeholders ensures this as it can provide a scope for independent visibility and recognition amongst peers. For example, in the Indian Genome Variation Consortium project which involved 200 participants, all the major stakeholders were from the same domain. The cohesiveness could not be sustained amongst stakeholders as once the dataset and platform were ready, everyone could independently take off to do solo projects.
Who leads in a collaboration?
A major aspect of the collaboration is the criteria for choosing the leader and also defining the leadership hierarchy in terms of reporting structure. An essential attribute in a leader besides a demonstrable competence is interpersonal communication skills, emotional intelligence and empathy. Also, the manner in which the leader handles conflicts, by consensus or autocratic methods, determines the cohesiveness within the team and his/her acceptance as a leader. Depending on the kind of collaboration, there might be requirements of multiple leaders from different domains. The roles of these leaders should be clearly defined. There is also a key role for mentorship and monitoring needs in major collaborative efforts. Choosing the appropriate mentors should be one of the critical aspects in such initiatives. The mentors should not have biased or personal interests or stakes in the project. It is important to distinguish between mentor, leader and manager (discussed more separately). Oftentimes the lead PI is assumed to be all three.
Solo leaders directing a multi-disciplinary team where the subordinates are from diverse backgrounds can hamper progress especially if they comprise a major percent. In long-term collaborative projects another aspect to consider is the growth dynamics of the team players and also the mentor-mentee dynamics. A person who ideates the project needs to relinquish leadership at some point of time with the evolution of the project and growing experience of team members.
In the ataxia project there has been a succession of leadership: from a biophysicist to geneticist and now a clinician leading the project. It started with how repetitive sequences are involved in pathogenesis of ataxia by looking at natural variations associated with disease status in multi-generation families. As the registry built up, the question evolved to using the variations to trace founders and identify new variations linked to diseases. Further, induction of clinicians in the backdrop of a registry enabled linking genotypes to phenotypic trajectories and resolving unknown cases through next generation sequencing approaches. This project is now poised for a iPSC approach, and expanding to include stem cell biologists. A glue that held the work together was the involvement of a single clinical investigator throughout the research period, who now heads a specialised ataxia clinic in AIIMS New Delhi. The sustenance of this program was due to successive leadership who had the expertise and interest in the project as it evolved organically.
What about team members?
The composition of the team could go a long way towards determining the success of the collaboration. As mentioned earlier, it should ensure adequate diversity so that contributions of the team players are evident. If all the team members have similar participation, then the collaborative members could become competitors eventually. Some key characteristics that determine team success include experience and expertise of team players, the alignment of individuals’ aspirations as well as their initial conditioning. Understanding of temperament, mindsets as well as value systems amongst the team members can minimise inter-individual conflicts. It is also very important that each of the team members are clear about their expertise and have their growth trajectories aligned to the collaboration for them to effectively contribute to the team efforts.
The Ayurgenomics project has had a diverse set of people with expertise from very contrasting domains. It has been the most challenging and the most engaging project personally for me. However, since it required a major cross-talk between genomics, Ayurveda, modern medicine and computational sciences, a major effort was devoted to developing an effective dialogue for communications amongst the researchers. The language, conditioning and approaches in each of these disciplines is very different and there are limited avenues where cross-talks between these disciplines are possible. This is compounded by the lack of funding organisations that provide a level playing field for researchers who are from different philosophies. The incentive framework and appraisal mechanisms are benchmarked to whatever is mainstream. This cannot create a sustainable framework for innovative research.
How do you share credits?
In large collaborative projects the credit sharing and attributions should be discussed right at the inception of the project else it becomes very ticklish at a later point and sometimes, professional relations get affected. In collaborative projects involving radical innovations success can take a longer time compared to one that involves incremental innovation. Some aspects of collaborative work cannot be completed unless initial frameworks are set. How do you credit the people who had been key during inception where the outcomes are not apparent? In these kinds of projects, it is important to keep a trail of contributions from inception to completion. There could be a few team members who play critical roles during inception and some during closures. If the gaps from inception to closure are very wide, the contributions of the initial players do not get adequately recognized or credited and the individuals who participate during closure of the projects share most of the limelight. There should also be attempts to define outcomes in definitive milestones to keep the team together. This also ensures that the contribution of maximum team members is visible. If independence and ownership for the sub-projects at all levels of functional hierarchy is provided, more people can be incentivized in collaborative projects. However, a balance of quality and quantity needs to be explicit.
One of the interesting things we initiated in the Indian Genome Variation Project was to have a review paper written up about the project with the attributions clearly mentioned. Subsequently any manuscript which solely used the resources had the consortium as one of the authors. To this date, Indian Genome Variation consortium is credited with an authorship in nearly 40 manuscripts and the same has been continued for TRISUTRA Ayurgenomics consortium and, also adopted in the PANASIAN SNP consortium.
Do we need managers?
An important cultural aspect, unfortunately prevalent in Indian science, is a poor respect for timelines and time of collaborative partners. A collaborative project would have complex dependencies at the organizational level, between team members and between organizations. This is where the role of managers is very crucial and we have to give due consideration to this critical manpower in major collaborative projects. Clearly defined milestones and definitive outcomes at each milestone are crucial to sustaining a team throughout collaboration and it is the role of the manager to ensure this. Anything managerial is anathema for a free-thinking scientist in an academic research, but this mindset needs to change.
There will always be many different ways in which we can pursue a fulfilling science career. Framing solo vs. collaborative research, merely in intellectual terms, has been counterproductive for Indian science. I hope that early career researchers find examples in this essay to encourage them to adopt a strategy that best fits with their research goals. Ultimately, it is “peers”, who create a scientific culture and the next generation may wish to do things differently.
The author is grateful to L S Shashidhara (IISER Pune, Ashoka University) and Megha (TDU) for providing critical comments and editorial support for this article.
About the AuthorProfessor Mitali Mukerji