Poster
Application of parallel machine learning methodologies to surface mechanistically distinct classes of compound hits in a high-content imaging screen for inflammasome inhibitors
November 1, 2023
•
7 minutes
Human innate immune responses critically rely on inflammasome pathways to convert internal and external danger stimuli into pro-inflammatory signals. Dysregulation of these pathways can lead to hyperinflammation and tissue damage, and inflammasome signaling has therefore been linked to the pathogenesis of a broad range of human conditions that vary from acute sepsis to chronic inflammation.
Inflammasome therapeutic molecule discovery typically relies on in vitro activation assays with terminal pathway readouts that are honed to identify robust, late-pathway inhibitors. These assays cannot distinguish between compound mechanisms of action, which limits optimal coupling of therapeutics to the wide array of inflammasome-related pathologies. We therefore conducted a high-content imaging screen of human PBMCs activated with two independent inflammasome stimuli and used ML (Machine Learning) analysis tools to identify and functionally define inflammasome inhibitors.
The following three complementary analytical approaches were applied to fluorescence imaging data with a goal of surfacing mechanistically distinct inflammasome inhibitors: (1) automated ASC speck quantification which was used as a ground truth for inflammasome activation, (2) targeted scoring that uses differentially weighted ML derived features associated with inflammasome activation and inhibition, (3) distance measurements in the unbiased phenotypic representation space.
933 compounds were identified as hits using the targeted scoring method. 807 of these (86.5%) were cross-validated using the distance method, while 850 compounds (91.9%) had measurable decreases in ASC specks under at least one of the two activation conditions tested. Altogether, there was strong agreement for hit designations between these methods which was further improved when more stringent scoring thresholds were applied (316 distance out of 338 targeted (93.5%); 322 of the 338 (95.3%) decreased ASC specks in at least one condition).
Identified compound hits were then categorized into functional classes using a set of scores that applied differential weight to biological and ML derived features associated with early stages of activation, reversible pathway convergence points, or the terminal, pyroptotic step where inflammasome pathways ultimately converge. The distance measurement was similarly used to resolve early and late inflammasome pathway inhibitors by setting distance thresholds based on the distribution of the control conditions. 91.4% of compounds (192/210) identified as late stage inhibitors using the terminal step scoring rubric were also identified as late inhibitors using the distance-based algorithm, indicating very close alignment and cross-validation between the two methods.
Altogether, we utilized ML and advanced multi-dimensional data analysis to surface and categorize mechanistically distinct classes of compound modulators from a high content imaging screen. These methods could be broadly applied to therapeutic compound screens or used to resolve complex, physiologically relevant biological pathways.
Rachel DeVay Jacobson*, Francesco Rubbo*, Lauren Nicolaisen, Daniel Chen, Michael Wiest, Adi Prakash, Will Van Trump, Tempest Plott, Lauren Nicolaisen, Ben Komalo, Elisa Cambronero, Christian Elabd
* denotes shared first authors