Explanation of output

[msingerlab]

Explanation of output

[Cnrdelaney]

Hello! Very happy to hear our tool has been useful. The columns are described as follows:

HG_pval - HyperGeometric p-value obtained from the hypergeometric test, this corresponds to statistical signifcance of the gene panel.
TP - True positive value, or the percentage of expressing cells in the cluster (using the given expression threshold).
TN - True negative value, or the percentage of non-expressing cells outside of the cluster (using the given expression threshold).
q_value - Q-value is an adjusted p-value
HG_rank - Ranking for the gene panel by lowest HG_pval
CCS - Cluster Clear Score, this is a value that assesses how much the additional gene (2- gene panels only) improves your True Negative score in other clusters. More about this can be read at our paper available on the bioarxiv (https://www.biorxiv.org/content/10.1101/655753v1)
MGD - Another way of scoring TN changes, not actually used at the moment
CCS_rank - Ranking of gene panel by highest CCS value
rank - Final rank of the panel, an aggregate of the HG_rank and the CCS_rank
Plot - 1 if the panel is plotted in the visualization files, 0 if it is not

In addition to this, there is some further description of the outputs on the documentation here (https://hgmd.readthedocs.io/en/latest/details.html#export-statistical-results)

Thanks and let me know if there is anything else confusing!