Academic 'disruptors' win start-up grants
Two projects aimed at disrupting the academic space have won Catalyst Grant awards for innovative startups.
Research industry technology company Digital Science has revealed that two US companies, Scismic and Rationally, have each been awarded a grant.
An international initiative to develop innovative projects and technologies The Catalyst Grant offers an award of up to £25,000 – or $30,000 – for concepts with the potential to transform scientific and academic research. The grant supports ideas at an early stage of development, without the need for a complete business or development plan.
Scismic Job Seeker is an online, diversity-promoting, automated recruiting platform for the biopharmaceutical industry. The platform matches scientists to jobs based on expertise and removes sources of bias, with its gender and race-blind matching algorithms, helping increase diversity in scientific hiring.
'Our goal is to help all scientists, no matter their background, find workplaces that empower them to propel ground-breaking science,' said co-founder Elizabeth Wu. 'One major barrier to scientific innovation is workforce development.'
Scismic was founded by three Boston-based scientists looking to address the $1B annual loss from inefficiencies in scientific recruiting in the US, with plans to further expand into the $8.5B STEM recruiting industry. Their motivations came from observing many talented colleagues, initially enthused with scientific drive, grow disengaged and unproductive in their jobs, feeling stuck in their careers.
'Eventually, the three of us went through career changes and personally experienced the challenges of transitioning from academia to industry,' said co-founder Danika Khong. They decided to build a platform, with support from entrepreneur Danny Gnaniah, that could address career challenges and recruiting inefficiencies.
'We wanted to build a way for fellow researchers to find workplaces in academia and industry that would empower them to do their best science, and drive more research into the market,' added Wu.
The co-founders highlight that current recruiting processes take over 2.5 months to fill vacancies and often results in poor fitting roles, lowering productivity and delaying scientific innovation. In addition, recruiting processes include sources of biases in candidate evaluation, like candidate name, which have been shown to exclude underrepresented scientists.
Rationally is an online platform that guides good research design according to published standards (e.g. CONSORT, PCORI) and reduces predictable sources of irreplicability. Founded by Kristin Lindquist, who has worked at technology startups and in product design for over 10 years, the platform aims to guide researchers in their efforts to design more replicable, feasible and less biased experiments.
Rationally initially started out as passion project with a very different premise – a crowd sourced site answering questions with scientific evidence. However, while interacting with researchers and being exposed to the problems of the replication crisis, Lindquist changed direction to focus on researcher tools.
Unreproducible science, Lindquist says, contributes to the estimated $200bn in wasted biomedical investments each year. The founder argues the 'publish or perish' incentivises researchers to find the interesting / anomalous over the uninteresting / replicable: 'Compounding the problem, good research design is hard, peer review is untimely, and meticulous science isn't readily distinguished from the sloppy. An epidemic of poor study design and R&D waste results. How do we get the eight million researchers in the world to know about and adhere to better practices?'
The replication crisis is particularly impactful in fields such as biomedical, psychology and sociology. Some 89 per cent of the foundational oncology studies evaluated by Amgen failed replication and an analysis by Bayer researchers revealed a mere 20 to 25 per cent replication success rate of the biomedical research investigated. A Nature survey asking 1,500 scientists what factors would boost reproducibility showed a better understanding of statistics, better mentoring and supervision and more robust design were key priority areas.
'What makes for good research design isn’t so much a problem of insights as it is of dissemination,' said Lindquist. 'The scientific community produces excellent guidance on research reproducibility, yet the replication crisis is still going strong. We’ve tried to reduce an ambiguous and complex process into a guided, step-by-step experience,' says Lindquist. 'It helps researchers think about how their study will be perceived in the meta-analysis process or by research reliability experts.'
The platform helps researchers to understand feasibility, rigour and generalisability trade-offs to make optimal design decisions. Currently in alpha, Rationally also helps connect researchers with biostatistics experts (at their institution or remote) for design review and help on areas of complexity. It can also generate checklists, methodology outlines or execution flow charts from research design specifications.
Steve Scott, director of portfolio development at Digital Science said: 'Twice a year we open the Catalyst Grant to applications from all over the world – and each year the number of entries continues to grow.
'The people best positioned to define areas for innovation are researchers themselves – but it’s incredibly hard for those with an idea to secure early-stage funding – finding investors who understand the research market is a challenge, meaning many potentially successful ideas remain just that, "ideas". That’s exactly why we created the Catalyst Grant – our financial support, alongside our advice makes a real difference.'
