Scientists at NCATS have developed a novel Cellular Thermal Shift Assay (CETSA), named “Real-time CETSA” in which temperature-induced aggregation of proteins can be monitored in cells in real time across a range of compound concentrations and simultaneously across a temperature gradient in a high-throughput manner. Real-time CETSA streamlines the thermal shift assay and allows investigators to capture full aggregation profiles for every sample. A traditional CETSA assay can only be run at a single temperature (dose-response compounds) or a single compound concentration (across temperature range). The real-time CETSA approach allows both variables to be captured in a single experiment in high throughput, as protein aggregation is monitored in real-time over a temperature gradient. The real-time method also allows multiple different proteins (e.g., multiple members of a target class) to be examined in parallel, even if they have vastly different aggregation profiles. Importantly, RT-CETSA can be measured in intact living cells as well as cell lysates.

Additionally, the scientists have developed a novel method of analyzing the RE-CETSA data that uses a nonparametric analysis of goodness of fit tests between linear and log-logistic models across the entire melting curve, in contrast to the current single-parameter analysis. While traditional methods of analysis for CETSA data rely on single parameter measurements, such as Tagg (temperature of aggregation, or the midpoint of the sigmoidal thermal unfolding curve) and AUC (area under the curve of a sigmoidal fit), such analyses do not take into account the entire thermal curve that RT-CETSA generates. Using a nonparametric analysis of goodness of fit tests between linear and log-logistic models across the entire melting curve results in more sensitive and reproducible hit identification and full use of the RT-CETSA data. There is no current analysis method for RT-CETSA, differential scanning fluorimetry (DSF), or other such methods, where there is integration of goodness of fit nonparametric testing with concentration-response data.