I hereby claim:
- I am dhimmel on github.
- I am dhimmel (https://keybase.io/dhimmel) on keybase.
- I have a public key whose fingerprint is 8CA9 B3A5 A4B4 D442 A285 867B 03A8 B7F8 47EE A45D
To claim this, I am signing this object:
Verifying that +dhimmel is my Bitcoin username. You can send me #bitcoin here: https://onename.io/dhimmel |
library(ROCR) | |
library(caTools) | |
VariableThresholdMetrics <- function(score, status) { | |
#' Evaluate the performance of predictions for a binary outcome. | |
#' | |
#' @param score A vector of predictions for which performance should be evalauted. | |
#' @param status A vector of the actual outcome: \code{0} (for negatives) and \code{1} (for positives). | |
#' @return A list. |
bto_id | bto_name | cell_line | |
---|---|---|---|
BTO:0000038 | SW-480 cell | 1 | |
BTO:0000041 | medulla oblongata | 0 | |
BTO:0000045 | adrenal cortex | 0 | |
BTO:0000047 | adrenal gland | 0 | |
BTO:0000084 | vermiform appendix | 0 | |
BTO:0000089 | blood | 0 | |
BTO:0000141 | bone marrow | 0 | |
BTO:0000142 | brain | 0 | |
BTO:0000211 | caudate nucleus | 0 |
# Compare four methods for computing the R-squared (R2, coefficient of determination) | |
# with wieghted observations for a linear regression model in R. | |
# This work was written by Daniel Himmelstein (@dhimmel) with guidance | |
# from Alex Pankov (@a-pankov). It is released as CC0 (public domain). | |
get_r2_cor <- function(y, y_pred, w) { | |
# Calculate R2 using the correlation coefficient method | |
xy = cbind(y, y_pred) | |
return(boot::corr(d=xy, w=w) ^ 2) | |
} |
I hereby claim:
To claim this, I am signing this object:
We designed a hetnet for drug repurposing that contains 50 thousand nodes (of 10 labels) and 3 million relationships (of 27 types). And we’ve chosen neo4j for handling network storage and interaction.