Northwestern maps protein landscapes

Northwestern University researchers are describing a new way to measure how proteins shift among different structural states, using an experimental setup built to work across many proteins at once rather than one by one. The work was detailed in a Nature paper published online on May 13 and highlighted by Northwestern Medicine in a May 21 news release. On June 1, a Northwestern-affiliated social post recirculated the result and pointed readers to the paper and supporting materials. The researchers said the approach could make it easier to compare how sequence changes alter protein behavior across large sets of molecules. ### What exactly did the Northwestern team build? The study introduced what the authors called a multiplexed experimental approach using intact-protein hydrogen-deuterium exchange mass spectrometry, or mHDX-MS. In the paper’s abstract, the team said the method analyzes the energies of conformational fluctuations for hundreds of protein domains in parallel. Northwestern Medicine’s release described it as a way to study conformational fluctuations “on a uniquely large scale.” (pubmed.ncbi.nlm.nih.gov) Gabriel Rocklin, an assistant professor of pharmacology at Northwestern University Feinberg School of Medicine, was the senior author. Állan Ramos Ferrari, a research assistant professor of pharmacology, was the lead author, according to the release and the paper metadata. ### What is a protein energy landscape in plain terms? (pubmed.ncbi.nlm.nih.gov) Proteins do not sit in one fixed shape. The Nature abstract says folded proteins continuously fluctuate between low-energy native structures and higher-energy conformations that can be partially or fully unfolded. Those rarer states can affect function, interactions, aggregation and immunogenicity, the authors wrote. Rocklin said in Northwestern’s release that researchers usually do not know the energies of those different conformations, or how rare or common they are, for most proteins. (news.feinberg.northwestern.edu) Ferrari said existing computational tools can model protein behavior, but experimental tools have lagged in measuring those behaviors at scale. (pubmed.ncbi.nlm.nih.gov) ### How big was the experiment? The paper said the researchers analyzed 5,778 protein domains, each 28 to 64 amino acids in length. The authors wrote that the measurements revealed hidden variation in conformational fluctuations, including among sequences that shared the same fold and overall folding stability. Northwestern’s release said older approaches generally let scientists study one protein at a time, while the new strategy was built for parallel analysis of hundreds of protein domains. (news.feinberg.northwestern.edu) Rocklin Lab’s website similarly says the group developed an approach to characterize hundreds of protein energy landscapes simultaneously using pooled protein expression and hydrogen exchange mass spectrometry. (pubmed.ncbi.nlm.nih.gov) ### Why were these states hard to capture before? Northwestern Medicine said high-energy states are rare and transient, making them difficult to study with current experimental methods. The paper adds that, while native protein structures can often now be predicted accurately, conformational fluctuations and their energies remain comparatively invisible and hard to predict, and large-scale measurements have been limited by experimental challenges. (news.feinberg.northwestern.edu) The same release said even small sequence changes from mutation or protein engineering can produce large changes in the populations of different conformational states. That is one reason the team framed scale as important: broader measurements make it possible to compare many related proteins under the same experimental scheme. (news.feinberg.northwestern.edu) ### What did the researchers say this is for? Northwestern Medicine said the method may improve data-driven modeling, biology and protein engineering. Rocklin Lab says its broader goal is to define quantitatively how protein energy landscapes arise from sequence and structure and to feed those data into predictive models for protein design. (news.feinberg.northwestern.edu) The June 1 social post pointed readers to the published work and related materials. The paper is listed in Nature with DOI 10.1038/s41586-026-10465-z, and Northwestern’s release identifies Rocklin and Ferrari as the key named participants in the next phase of follow-on work using the method. (pubmed.ncbi.nlm.nih.gov) (news.feinberg.northwestern.edu)