Cell signaling pathway discovery

We developed Incytr, a method that enables efficient discovery of cell signaling pathways by integrating diverse data modalities, including transcriptomics, proteomics, phosphoproteomics, and kinomics

We present an alternative clustering approach that mitigates the curse of dimensionality by sequentially projecting high-dimensional data into low-dimensional representations

We evaluated two prevailing approaches in the field: modeling TCR specificity and T cell activation as separate tasks, and using unsupervised, sequence similarity–based models to predict TCR specificity

We evaluated the generalization capacity of TCR specificity prediction models using a proprietary cancer-patient dataset with novel peptide sequences. Our findings reveal that while existing ML models perform well on known peptides, they fail to generalize to novel ones, where physics-based models perform better

We developed SPEX (Spatial Expression Explorer), a web-based platform featuring a modular analysis pipeline and a user-friendly interface

We present a flow cytometry data analysis pipeline that incorporates Earth Mover’s Distance (EMD) to enable statistically robust comparisons of biomarker expression across subject groups. Through three case studies, we demonstrate EMD’s ability to detect clinically and biologically meaningful differences that standard approaches may miss