Your body contains thousands of genes, but 90 per cent of biomedical research focuses on just 10 per cent of them.
Many of these neglected genes play a role in serious disease including breast cancer, so why aren’t they being studied?
The reasons are often logistical and mundane, according to a new research paper aiming to quantify why biomedical scientists keep returning to the familiar.
“To our surprise, we found out very generic chemical characteristics of proteins and genes account for much of the research which is currently done,” said lead researcher Thomas Stoeger.
The study, published in PLOS Biology, created a large database that cross-referenced data on all human protein-coding genes — including chemical, physical, historical and experimental data.
It then used machine learning to examine the relationship between a gene’s characteristics and how likely it was be studied.
Why some genes and not others?
In short, some genes aren’t studied because they are a little bit complicated.
If the protein made by a gene is big, that makes it hard to study.
If the protein the gene makes is kept inside the nucleus of a cell rather than secreted, that also makes it hard to study.
And researchers are far more likely to study genes we can easily model in animals, such as mice or fruit flies.
The study also shows that researchers are spending the most time looking at genes that were reachable in the 1980s and 1990s, when they were first identified.
“We find that despite technology being far more advanced nowadays, we do not apply this technology to new genes, but kind of stick with what we have been doing some decades ago,” Dr Stoeger said.
To be sure, the genes that account for the most research are more likely than your average gene to be important for disease. But the research effort is still disproportionate.
“While they’re roughly six times as likely to be important, there are roughly 8,000 times more publications,” Dr Stoeger said.
Staying on familiar paths
This bias breeds more bias, as early-career researchers play it safe.
Dr Stoeger and his team found PhD students and postdoctoral fellows who examined little-studied genes had a lower probability of becoming a principal investigator or of setting up their own research group, when compared to people who stuck to the well-trodden path.
The human genome was mapped at the turn of the millennium, but we’re still mostly studying genes we discovered decades ago — possibly at the expense of breakthroughs in understanding disease.
While Dr Stoeger’s study looked at data from the United States, this is a global research problem and Australia is no exception, said Darren Saunders, a cancer biologist at the University of New South Wales.
“It’s kind of a ‘rich gets richer’ principle at play here,” Dr Saunders said.
“The genes that people started working on early on, that were obviously identified as being important, once there’s a body of knowledge around those genes, people tend to gravitate towards working on them.
Working on a relatively unknown gene means taking the time to build the tools required to study that gene, which is tough in an industry where the currency is published research.
“By that time, a few years might have passed and if you haven’t been very productive in terms of publishing papers and winning grants, that tends to put a very big chilling effect on your career,” Dr Saunders said.
While the imbalance between the pursuit of truly novel medical breakthroughs and the logistical realities of research is well known within science, Dr Saunders acknowledged it could come as a shock to the rest of us who trust science to solve the big problems.
“There’s a big disconnect between how people think science works and how it actually works in practice,” he said.
How to get off the beaten research track
Correcting this imbalance is trickier than it seems, Dr Stoeger said.
In the United States, the National Institutes of Health set up a number of grant initiatives a few years ago to support exploratory research.
“But … despite the best intentions, these agendas still end up supporting research that’s very similar to other grant systems,” he said.
This new study offers some ideas on where to start, though.
The researchers used the same machine learning tool that helped them predict characteristics of past research to identify neglected genes that hold the strongest potential for new research.
Dr Darren Saunders says fellow researchers, not just funding bodies, need to shift their thinking.
ABC News: David Lewis
Here in Australia, the National Health and Medical Research Council (NHMRC) recently reformed its grant program to “encourage greater creativity” and avoid an environment that “favour[s] ‘safe’ research to the detriment of innovation”.
Dr Saunders said the changes were a good start, but a shift needs to happen at all levels.
“It’s up to all of us, both the people writing the grant funding and asking the questions in science, and those of us who review the funding applications, to try and be a little bit more risky, less risk-averse, in our approach to what we fund,” he said.
“Maybe we need to be a little bit more accommodating of scientists who take risks that don’t necessarily always have those risks pay off.
“Because that’s kind of the nature of science, right? You can’t always predict where it’s going to go.”