We want something like the bar charts generated in LuCA. Statistics need to be performed on the level of biological replicates.
def shears_stats(adata_sc, groupby="cell_type", batch_key=["patient"]):
passTo this end, we need to compute a score based on the pvalue and coefficient for each cell. Either by using a discrete cutoff (e.g. FDR < 0.01), or (my preferred id) to take a weighted sum of the coefficients per cell-type cluster, where the coefficients are weighted by the -log10(pvalue).
Together with the tool that generates a dataframe or similar, we also need a plotting function to generate e.g. a bar chart, or a pair-plot.
We want something like the bar charts generated in LuCA. Statistics need to be performed on the level of biological replicates.
To this end, we need to compute a score based on the pvalue and coefficient for each cell. Either by using a discrete cutoff (e.g. FDR < 0.01), or (my preferred id) to take a weighted sum of the coefficients per cell-type cluster, where the coefficients are weighted by the
-log10(pvalue).Together with the tool that generates a dataframe or similar, we also need a plotting function to generate e.g. a bar chart, or a pair-plot.