Doaa M. Taha

The paper introduces SI-SimCLR, a self-supervised learning framework for reconstructing cellular phenotypic landscapes from microscopy data without segmentation or annotation, revealing substate organization and connections in astrocytes.

The paper develops an automated image processing pipeline with machine learning to classify motor neuron phenotypes in ALS tissue at single-cell resolution, revealing common morphological aberrations and validating their discriminative power between ALS and healthy tissue.

The paper identifies aberrant cytoplasmic intron-retaining transcripts (IRTs) in ALS and shows their association with RBP mislocalization.

The paper investigates FUS mislocalization in ALS, finding it occurs in both VCP-mutation related and sporadic ALS, and links it to aberrant intron retention in SFPQ transcripts.

The paper identifies post-transcriptional changes during motor neurogenesis in ALS models, showing premature intron retention in VCP-mutant cells and similar effects in SOD1 and FUS mutations.