sim_score_grp {signatureSearch} | R Documentation |
Function appends two columns (score_column_grp1, score_column_grp2) to GESS
result tibble. The appended columns contain cell-summarized scores for
groups of cell types, such as normal and tumor cells. The cell-summarized
score is obtained the same way as the NCSct
scores, that is using a
maximum quantile statistic. It compares the 67 and 33 quantiles of scores
and retains whichever is of higher absolute magnitude.
sim_score_grp(tib, grp1, grp2, score_column)
tib |
tibble in gessResult object |
grp1 |
character vector, group 1 of cell types, e.g., tumor cell types |
grp2 |
character vector, group 2 of cell types, e.g., normal cell types |
score_column |
character, column name of similarity scores to be grouped |
tibble
gr <- gessResult(result=dplyr::tibble(pert=c("p1", "p1", "p2", "p3"), cell=c("MCF7", "SKB", "MCF7", "SKB"), type=rep("trt_cp", 4), NCS=c(1.2, 1, 0.9, 0.6)), query=list(up="a", down="b"), gess_method="LINCS", refdb="path/to/refdb") df <- sim_score_grp(result(gr), grp1="SKB", grp2="MCF7", "NCS")