When Sergey Frolov, a professor in the Department of Physics at the University of Pittsburgh’s Kenneth P. Dietrich School of Arts and Sciences, was asked by a collaborator to produce 15-year-old research data, he quickly located and shared the file within 20 minutes. This practice reflects an approach he now encourages among his students: keeping data organized, backed up, and readily available for sharing.
Frolov has extended this philosophy to his broader field in condensed matter physics. In a recent paper published in Science on January 8, he advocates for open data sharing and increased transparency as ways to address what is known as the replication crisis—a challenge affecting many areas of science where key findings cannot be reproduced by other researchers.
“Some theoretical predictions can bias you as a scientist,” Frolov said. “You see a pattern that you’re looking for, you can even convince yourself, ‘This must be it.’”
He cautions against relying on so-called “smoking gun” data—single pieces of evidence that appear to prove a phenomenon conclusively. “When it comes to making a scientific discovery, a smoking gun is a piece of data that contains the full proof of the phenomenon,” Frolov explained. “It’s a single figure that tells the entire story.” However, he notes such figures can sometimes mislead.
Frolov’s work centers on new states of matter relevant to quantum computing. In 2018, Microsoft researchers claimed they had discovered one such state—the Majorana fermion—presenting what appeared to be definitive evidence. Frolov requested additional unpublished data from their experiment and identified issues including selective use of non-representative results and potential manipulation. The original authors eventually retracted their paper after these concerns came to light.
Following this experience, Frolov and his team—including graduate students Yifan Jiang, Bomin Zhang and Seth Byard, as well as postdoctoral researcher Po Zhang—attempted to replicate four prominent experiments published with similarly dramatic results. Each time they found alternative explanations rather than confirmation. Submissions describing these replications were initially rejected by journals on grounds that replication studies lacked novelty or relevance due to elapsed time since original publication.
Persisting in their efforts, the team consolidated their analyses into one paper highlighting how extraordinary results often diminish when examined alongside complete datasets rather than select figures alone. This work was ultimately accepted by Science two years after initial submission.
In addition to urging more comprehensive reporting from researchers—including disclosure of alternative interpretations—Frolov points out that some disciplines have already adopted more transparent practices regarding data sharing; for example, many astrophysicists routinely share their experimental results openly.
He also suggests journal editors could provide readers with more insight into editorial decision-making processes surrounding publications—a step toward greater openness about which findings are considered robust enough for dissemination.
In 2024, Frolov collaborated with colleagues from several universities to organize a conference at the University of Pittsburgh focused on reproducibility in physics research; outcomes included both a detailed report and an educational video intended to foster support for change across the discipline.
“I’d be a fool if I expected a sudden change,” Frolov said. “There will be talking, then some changes, then there will be some backlash because changes cause irritation. But then there will be some adaptation, and there will be more talking.”
Ultimately, Frolov hopes these efforts will lead toward improved openness and honesty within scientific research practices.
