- Research collaboration calls for bioinformatics training boost at MSIs.
- NIH-funded MARC program aims to broaden access to biomedical training.
- Study shows multidisciplinary research is largely absent at MSIs.
Bringing more minority students into the bioinformatics training pipeline may require more funding and a major rethink of STEM outreach to minority-serving institutions (MSIs).
That’s the conclusion reached by a team from the University of Puerto Rico (UPR) and the Pittsburgh Supercomputing Center (PSC). The team surveyed students from MSIs before and after an intensive two-month summer bioinformatics internship offered by PSC, and presented their research at the XSEDE16 supercomputing conference in Miami in July.
PSC, in collaboration with MSI partners, developed the Minority Access to Research Careers (MARC) program to help increase minority participation in biomedical research.
Funded by the US National Institute of Health (NIH), MARC included a seven-week summer research internship at PSC for students who’ve had bioinformatics training on their local campus.
MARC also included the development of a model curriculum for bioinformatics and assistance to MSIs in establishing and strengthening bioinformatics programs at two MSI campuses each year.
“A large number of schools, students and faculty do not have the computational or bioinformatics skills necessary to carry out competitive biomedical genomics research,” observes Ricardo Gonzalez Mendez of the UPR Medical Sciences. “We need to increase the amount of instruction that builds analytical, computational and Next-Generation Sequencing (NGS) analysis skills to train novices in these areas.”
The MARC summer internship raised participating students’ bioinformatics skills across the board, Gonzalez Mendez notes. Telling, however, was the students’ lack of previous exposure to basic bioinformatics skills.
Despite being among the top students at their institutions, 27 percent ranked at the ‘novice’ level in performing sequence database searches or multiple sequence alignments; 59 percent were ranked no higher than a ‘basic’ level of skill in either. For a set of nine specialized NGS skills surveyed, between 68 percent and 91 percent ranked only as novices.
“Most of these students still think, ‘I’m studying biology,’ or ‘I’m studying computer science.’ But you need to think about combining these skillsets for a successful career in the 21st century.” ~ Pallavi Ishwad
These skills are important, the researchers note, because having exposure to them is necessary to derive benefit from the typical one- to two-day outreach program, let alone embark on successful bioinformatics study.
“Our whole mindset needs to change,” says coauthor Pallavi Ishwad, education program director at PSC. “Most of these students still think, ‘I’m studying biology,’ or ‘I’m studying computer science.’ But you need to think about combining these skillsets for a successful career in the 21st century.”
As part of their standard curricula, top-100-funded colleges and universities have largely made the transition to teaching bioinformatics, says coauthor Alexander Ropelewski, director of PSC’s Biomedical Applications Group. But such multidisciplinary coursework, particularly early in the curricula, is largely absent at MSIs. As these institutions educate some 58 percent of minority students, this represents a major obstacle for minority communities.
One reason for the gap between the bioinformatics ‘have’ and ‘have not’ institutions stems from a self-sustaining funding gap. Lacking large research grants, MSIs are funded mostly by student tuition. This in turn leads to high teaching loads for faculty, who then have less time for research and writing research grants.
“What is needed is much more intense training that gets the MSI faculty to the point that they can effectively transition the coursework to emphasize building the quantitative skills needed by the biomedical workforce,” Ropelewski says. “Short courses can be successful with an audience that’s already had a lot of the prerequisites.”
The short-course approach has worked well in PSC’s and XSEDE’s supercomputing programming workshops for the general scientific community, he adds. But it may not be up to the task of addressing the needs of students without background knowledge.
“We need to create a path to help those individuals get the requisite knowledge, skills or whatever they need,” he adds. “Completing a bioinformatics project should really be the goal.”