DestVI

scRNA-seq is generated based on learn NB parameters from the destVI manuscripts leveraging sparsePCA. Number of cells and cell types present in each spatial spot is computed via combination of kernel-based parametrization of a categorical distribution and the NB model.

DestVI

RankMean scorer2Memory (GB)Runtime (min)CPU (%)NamePaperYearLibrary
10.900.9040.30205.732,109.80Cell2location, amortised (detection_alpha=20, reference hard-coded)2022v0.1
20.900.903.10128.481,279.60Cell2location (detection_alpha=200, reference hard-coded)2022v0.1
30.900.903.10101.531,220.90Cell2location (detection_alpha=20, reference hard-coded)2022v0.1
40.900.903.90114.581,214.80Cell2location (detection_alpha=20, reference NB without batch info)2022v0.1
50.740.744.0026.52100.70RCTD2020v2.0.1
60.490.490.480.3265.50Non-Negative Least Squares1987v1.9.3
70.400.402.303.05982.20NMF-reg2019v1.1.3
80.330.331.5029.231,649.20Stereoscope2020v0.16.4
90.320.321.300.76887.10Non-Negative Matrix Factorization (NMF)2009v1.1.3
10-0.04-0.042.2030.421,573.70DestVI2022v0.16.4
11-0.95-0.953.2022.37746.70Tangram2021v1.0.3
12-2.58-2.58211.209.62120.80SeuratV32019v4.1.1