Update pseudobulk.R

}

#' @export

collapse_counts <- function(counts_mat, meta_data, varnames, min_cells_per_group=0, keep_n=FALSE, how=c('sum', 'mean')[1]) {

    ## give each unique row a hash value for indexing

collapse_counts <- function (counts_mat, meta_data, varnames, min_cells_per_group=0, 

    keep_n=TRUE, how = c("sum", "mean")[1], get_nnz=FALSE, get_norm=TRUE) 

{

    res <- list()

    hash <- compute_hash(meta_data, varnames)

    idx_keep <- which(!is.na(hash))

    hash <- hash[idx_keep]

    hash <- factor(sprintf('sample_%d', as.integer(hash)))

    hash <- factor(sprintf("sample_%d", as.integer(hash)))

    meta_data <- meta_data[idx_keep, ]

    counts_mat <- counts_mat[, idx_keep]

    ## one hot encoded design matrix, sample level

    design_collapsed <- data.frame(meta_data)[, varnames, drop = FALSE] %>% 

        cbind(sample_id = hash) %>% 

        unique()

    design_collapsed <- data.table(meta_data)[

        , varnames, drop = FALSE, with = FALSE

    ][

        , sample_id := hash

    ][

        , N := .N, by = sample_id

    ][

        N >= min_cells_per_group

    ] %>% 

    unique() %>% 

        cbind(sample_id = hash) %>% unique()

    design_collapsed <- data.table(meta_data)[, varnames, drop = FALSE, 

        with = FALSE][, `:=`(sample_id, hash)][, `:=`(N, .N), 

        by = sample_id][N >= min_cells_per_group] %>% unique() %>% 

        data.frame()

    ## sum over samples

    counts_collapsed <- presto:::sumGroups(counts_mat, hash, 1) %>% t()

    counts_collapsed <- presto:::sumGroups(counts_mat, hash, 

        1) %>% t()

    row.names(counts_collapsed) <- row.names(counts_mat)

    colnames(counts_collapsed) <- levels(hash)

    ## reorder to match design matrix

    counts_collapsed <- counts_collapsed[, design_collapsed$sample_id]

    design_collapsed$sample_id <- NULL

    if (how == 'mean') {

    if (how == "mean") {

        counts_collapsed <- as.matrix(counts_collapsed %*% Matrix::Diagonal(x = 1/design_collapsed$N))

    }

    if (!keep_n) {

        design_collapsed <- dplyr::select(design_collapsed, -N)

    }

    if (get_nnz) {

        pct_nnz <- presto:::nnzeroGroups.dgCMatrix(counts_mat, hash, MARGIN = 1)

        pct_nnz <- Diagonal(x = 1 / table(hash)) %*% pct_nnz

        # pct_nnz <- data.frame(t(as(pct_nnz, 'matrix'))) 

        pct_nnz <- Matrix::t(pct_nnz)

        rownames(pct_nnz) <- rownames(counts_mat)

        colnames(pct_nnz) <- levels(hash)

        res$pct_nnz <- pct_nnz[, design_collapsed$sample_id]

    }

    row.names(design_collapsed) <- design_collapsed$sample_id

    return(list(counts_mat = counts_collapsed, meta_data = design_collapsed))

    design_collapsed$sample_id <- NULL

    res$counts_mat <- counts_collapsed

    res$meta_data <- design_collapsed

    if (get_norm) {

        message('CAREFUL: get_norm makes very strong assumptions about data')

        res$exprs_norm <- normalizeData(res$counts_mat, 1e4, 'log')

        res$meta_data$logUMI <- log(colSums(res$counts_mat))

    }

    return(res)

}

# #' @export

# pseudobulk_pairwise <- function(dge_formula, counts_df, meta_data, contrast_var, vals_test, verbose,