proteinIDE - a low-compute, accessible way to predict proteins using quantization and disk-caching
A protein folding IDE with custom mutation prediction and real-time 3D rendering. Still a work in progress.
PSFs (or the lack thereof) - a fix to unrealistic degredation modeling for different microscope modalities
Building a physics-informed PSF model using physics priors for microscopy deconvolution. Accepted by CVPR.
A Better UMAP - self explanatory i think
Addressing UMAP's unreliability for single-cell data visualization. Building a better dimensionality reduction approach.