R/helper.R

In wefang/riso: Epigenome Isoform Analysis

# calProb <- function(mat, p, q, bg.mean, bg.sd, theta1, sigma1){
# 
#     if(is.vector(mat)) mat <- matrix(mat, 1)
#     if(is.vector(bg.mean)) bg.mean <- matrix(bg.mean, 1)
#     if(is.vector(bg.sd)) bg.sd <- matrix(bg.sd, 1)
# 
#     I <- dim(mat)[1]
#     J <- dim(mat)[2]
#     K <- dim(q)[1]
# 
#     like1 <- like0 <- matrix(, I, J)
#     temp.like.sum <- temp.like.ratio <- array(, c(I, J, K))
# 
#     for (i in 1:I){
#         for (j in 1:J){
#             like0[i, j] <- dnorm(mat[i, j], bg.mean[i, j], bg.sd[i, j])
#             like1[i, j] <- dnorm(mat[i, j], theta1[j], sigma1[j])
#         }
#     }
# 
#     for (i in 1:I){
#         for (k in 1:K){
#             for (j in 1:J){
#                 temp.like.sum[i, j, k] <- q[k, j] * like1[i, j] + (1 - q[k, j]) * like0[i, j]
#                 temp.like.ratio[i, j, k] <- q[k, j] * like1[i, j] /
#                     ( q[k, j] * like1[i, j] + (1 - q[k, j]) * like0[i, j] )
#             }
#         }
#     }
# 
#     clust.like <- matrix(, I, K)
#     all.like <- numeric(I)
# 
#     for (k in 1:K){
#         for (i in 1:I){
#             clust.like[i, k] <- log(p[k]) + sum(log(temp.like.sum[i, , k]))
#         }
#     }
# 
#     for (i in 1:I){
#         temp <- clust.like[i, ]
#         temp.max <- max(temp)
#         temp <- temp - temp.max
#         temp <- exp(temp)
#         clust.like[i, ] <- temp / sum(temp)
#         all.like[i] <- log(sum(temp)) + temp.max
#     }
#     loglike <- sum(all.like)
# 
#     cond.like <- array(, c(I, J, K))
#     for (i in 1:I){
#         for (k in 1:K){
#             for (j in 1:J){
#                 cond.like[i, j, k] <- temp.like.ratio[i, j, k] * clust.like[i, k]
#             }
#         }
#     }
# 
#     b.prob <- colMeans(clust.like)
#     a.prob <- Reduce("+", lapply(1:K, function(k) cond.like[, , k]))
# 
#     return(list(b.prob = b.prob, a.prob = a.prob, loglike = loglike))
# }

    # 
    # gatherPvalues <- function(da.out){
    #     do.call(rbind, lapply(da.out, function(out) {
    #                               if (class(out) == "try-error" | is.null(out$id)) {
    #                                   return(NULL)
    #                               } else {
    #                                   data.frame(id = out$id,
    #                                              isoform = out$isoform,
    #                                              pvalue = out$pvalue,
    #                                              statistics = out$test)
    #                               }
    # }))}