Computational models

A major challenge in analyzing scRNA-seq data arises from challenges related to dimensionality and the prevalence of dropout events. Here the authors develop a deep graph learning method called scMGCA based on a graph-embedding autoencoder that simultaneously learns cell-cell topology representation and cluster assignments, outperforming other state-of-the-art models across multiple platforms.

Trial data have shown that post-discharge malaria chemoprevention (PDMC) reduces the risk of readmission and death in children previously hospitalised with severe malarial anaemia. Here, the authors use mathematical modelling to estimate the potential epidemiological impacts of PDMC in malaria-endemic countries in Africa.

Many methods for single cell data integration have been developed, though mosaic integration remains challenging. Here the authors present scMoMaT, a mosaic integration method for single cell multi-modality data from multiple batches, that jointly learns cell representations and marker features across modalities for different cell clusters, to interpret the cell clusters from different modalities.