Abstract: Recent advances in sequencing technologies have alowed us to capture various aspects of the genome at single-cel resolution. However, except for a few co-assaying technologies, it is not possible to simultaneously apply different sequencing assays on the same single cel. In this scenario, computational integration of multi-omic measurements is crucial to enable joint analyses. This integration task is particularly chalenging due to the lack of sample-wise or feature-wise correspondences. In this talk, I wi l present our methods- Single-Cel alignment with OptimalTransport (SCOT) and Augmented Gromov Wasserstein Optimal Transport (AGWOT)- unsupervised algorithms that use optimal transport to align single-cel multi-omics datasets. I wi l discuss the interesting properties of these methods, such as automatic hyperparameter selection, extensions to the integration setting where datasets can have disproportionate cel type representations, and simultaneous alignment of cels and features for hypothesis generation in biology.

Reading list:

• [1] Liu, J., Huang, Y., Singh, R., Vert, J.P. and Noble, W.S., 2019, September. Jointly embedding multiple single-cel omics measurements. In Algorithms in bioinformatics:... International Workshop, WABI..., proceedings. WABI (Workshop) (Vol. 143). NIH Public Access.
• [2] Demetci, P., Santorela, R., Sandstede, B., Noble, W.S. and Singh, R., 2022. SCOT: single-ce l multi-omics alignment with optimal transport. Journal of computational biology, 29(1), pp.3-18.