Implement all methods using positions additionally

R/analyze.R

@@ -35,7 +35,19 @@ analyze <- function(preset, progress = NULL) {
         "clusteriness" = clusteriness,
         "correlation" = correlation,
         "proximity" = proximity,
-        "neural" = neural
+        "neural" = neural,
+        "clusteriness_positions" = function(...) {
+            clusteriness(..., use_positions = TRUE)
+        },
+        "correlation_positions" = function(...) {
+            correlation(..., use_positions = TRUE)
+        },
+        "proximity_positions" = function(...) {
+            proximity(..., use_positions = TRUE)
+        },
+        "neural_positions" = function(...) {
+            neural(..., use_positions = TRUE)
+        }
     )
 
     results <- cached("analysis", preset, {
@@ -50,7 +62,11 @@ analyze <- function(preset, progress = NULL) {
                 }
             }
 
-            method_results <- methods[[method_id]](preset, method_progress)
+            method_results <- methods[[method_id]](
+                preset,
+                progress = method_progress
+            )
+
             setnames(method_results, "score", method_id)
 
             results <- merge(

R/clusteriness.R

@@ -36,11 +36,11 @@ clusteriness_priv <- function(data, height = 1000000) {
 }
 
 # Process genes clustering their distance to telomeres.
-clusteriness <- function(preset, progress = NULL) {
+clusteriness <- function(preset, use_positions = FALSE, progress = NULL) {
     species_ids <- preset$species_ids
     gene_ids <- preset$gene_ids
 
-    cached("clusteriness", c(species_ids, gene_ids), {
+    cached("clusteriness", c(species_ids, gene_ids, use_positions), {
         results <- data.table(gene = gene_ids)
 
         # Prefilter the input data by species.
@@ -54,7 +54,13 @@ clusteriness <- function(preset, progress = NULL) {
 
         # Perform the cluster analysis for one gene.
         compute <- function(gene_id) {
-            score <- clusteriness_priv(distances[gene_id, distance])
+            data <- if (use_positions) {
+                distances[gene_id, position]
+            } else {
+                distances[gene_id, distance]
+            }
+
+            score <- clusteriness_priv(data)
 
             if (!is.null(progress)) {
                 genes_done <<- genes_done + 1

R/correlation.R

@@ -1,30 +1,40 @@
 # Compute the mean correlation coefficient comparing gene distances with a set
 # of reference genes.
-correlation <- function(preset, progress = NULL) {
+correlation <- function(preset, use_positions = FALSE, progress = NULL) {
     species_ids <- preset$species_ids
     gene_ids <- preset$gene_ids
     reference_gene_ids <- preset$reference_gene_ids
 
-    cached("correlation", c(species_ids, gene_ids, reference_gene_ids), {
+    cached(
+        "correlation",
+        c(species_ids, gene_ids, reference_gene_ids, use_positions),
+        { # nolint
             # Prefilter distances by species.
             distances <- geposan::distances[species %chin% species_ids]
 
             # Tranform data to get species as rows and genes as columns. We
-        # construct columns per species, because it requires fewer iterations,
+            # construct columns per species, because it requires fewer
-        # and transpose the table afterwards.
+            # iterations, and transpose the table afterwards.
 
             data <- data.table(gene = gene_ids)
 
             # Make a column containing distance data for each species.
             for (species_id in species_ids) {
-            species_distances <- distances[
+                species_data <- if (use_positions) {
+                    setnames(distances[
+                        species == species_id,
+                        .(gene, position)
+                    ], "position", "distance")
+                } else {
+                    distances[
                         species == species_id,
                         .(gene, distance)
                     ]
-
-            data <- merge(data, species_distances, all.x = TRUE)
-            setnames(data, "distance", species_id)
                 }
+            }
+
+            data <- merge(data, species_data, all.x = TRUE)
+            setnames(data, "distance", species_id)
 
             # Transpose to the desired format.
             data <- transpose(data, make.names = "gene")
@@ -54,8 +64,8 @@ correlation <- function(preset, progress = NULL) {
             if (!is.null(progress)) progress(0.66)
 
             # Compute the final score as the mean of known correlation scores.
-        # Negative correlations will correctly lessen the score, which will be
+            # Negative correlations will correctly lessen the score, which will
-        # clamped to zero as its lower bound. Genes with no possible
+            # be clamped to zero as its lower bound. Genes with no possible
             # correlations at all will be assumed to have a score of 0.0.
 
             compute_score <- function(scores) {
@@ -75,5 +85,6 @@ correlation <- function(preset, progress = NULL) {
             ]
 
             results[, .(gene, score)]
-    })
+        }
+    )
 }

R/neural.R

@@ -1,46 +1,62 @@
 # Find genes by training a neural network on reference position data.
 #
 # @param seed A seed to get reproducible results.
-neural <- function(preset, progress = NULL, seed = 448077) {
+neural <- function(preset,
+                   use_positions = FALSE,
+                   progress = NULL,
+                   seed = 448077) {
     species_ids <- preset$species_ids
     gene_ids <- preset$gene_ids
     reference_gene_ids <- preset$reference_gene_ids
 
-    cached("neural", c(species_ids, gene_ids, reference_gene_ids), {
+    cached(
+        "neural",
+        c(species_ids, gene_ids, reference_gene_ids, use_positions),
+        { # nolint
             set.seed(seed)
             gene_count <- length(gene_ids)
 
             # Prefilter distances by species.
             distances <- geposan::distances[species %chin% species_ids]
 
-        # Input data for the network. This contains the gene ID as an identifier
+            # Input data for the network. This contains the gene ID as an
-        # as well as the per-species gene distances as input variables.
+            # identifier as well as the per-species gene distances as input
+            # variables.
             data <- data.table(gene = gene_ids)
 
-        # Buffer to keep track of species included in the computation. Species
+            # Buffer to keep track of species included in the computation.
-        # from `species_ids` may be excluded if they don't have enough data.
+            # Species from `species_ids` may be excluded if they don't have
+            # enough data.
             species_ids_included <- NULL
 
             # Make a column containing distance data for each species.
             for (species_id in species_ids) {
-            species_distances <- distances[
+                species_data <- if (use_positions) {
+                    setnames(distances[
+                        species == species_id,
+                        .(gene, position)
+                    ], "position", "distance")
+                } else {
+                    distances[
                         species == species_id,
                         .(gene, distance)
                     ]
+                }
 
                 # Only include species with at least 25% known values.
 
-            species_distances <- stats::na.omit(species_distances)
+                species_distances <- stats::na.omit(species_data)
 
                 if (nrow(species_distances) >= 0.25 * gene_count) {
                     species_ids_included <- c(species_ids_included, species_id)
                     data <- merge(data, species_distances, all.x = TRUE)
 
                     # Replace missing data with mean values. The neural network
-                # can't handle NAs in a meaningful way. Choosing extreme values
+                    # can't handle NAs in a meaningful way. Choosing extreme
-                # here would result in heavily biased results. Therefore, the
+                    # values here would result in heavily biased results.
-                # mean value is chosen as a compromise. However, this will of
+                    # Therefore, the mean value is chosen as a compromise.
-                # course lessen the significance of the results.
+                    # However, this will of course lessen the significance of
+                    # the results.
 
                     mean_distance <- round(species_distances[, mean(distance)])
                     data[is.na(distance), distance := mean_distance]
@@ -56,10 +72,10 @@ neural <- function(preset, progress = NULL, seed = 448077) {
             reference_data[, neural := 1.0]
 
             # Take out random samples from the remaining genes. This is another
-        # compromise with a negative impact on significance. Because there is
+            # compromise with a negative impact on significance. Because there
-        # no information on genes with are explicitely *not* TPE-OLD genes, we
+            # is no information on genes with are explicitely *not* TPE-OLD
-        # have to assume that a random sample of genes has a low probability of
+            # genes, we have to assume that a random sample of genes has a low
-        # including TPE-OLD genes.
+            # probability of including TPE-OLD genes.
 
             without_reference_data <- data[!gene %chin% reference_gene_ids]
 
@@ -72,8 +88,8 @@ neural <- function(preset, progress = NULL, seed = 448077) {
 
             reference_samples[, neural := 0.0]
 
-        # Merge training data. The training data includes all reference genes as
+            # Merge training data. The training data includes all reference
-        # well as an equal number of random sample genes.
+            # genes as well as an equal number of random sample genes.
             training_data <- rbindlist(list(reference_data, reference_samples))
 
             # Construct and train the neural network.
@@ -96,8 +112,8 @@ neural <- function(preset, progress = NULL, seed = 448077) {
 
             if (!is.null(progress)) {
                 # We do everything in one go, so it's not possible to report
-            # detailed progress information. As the method is relatively quick,
+                # detailed progress information. As the method is relatively
-            # this should not be a problem.
+                # quick, this should not be a problem.
                 progress(0.5)
             }
 
@@ -110,5 +126,6 @@ neural <- function(preset, progress = NULL, seed = 448077) {
             }
 
             data[, .(gene, score)]
-    })
+        }
+    )
 }

R/proximity.R

@@ -2,21 +2,26 @@
 #
 # A score will be given to each gene such that 0.0 corresponds to the maximal
 # mean distance across all genes and 1.0 corresponds to a distance of 0.
-proximity <- function(preset, progress = NULL) {
+proximity <- function(preset, use_positions = FALSE, progress = NULL) {
     species_ids <- preset$species_ids
     gene_ids <- preset$gene_ids
 
-    cached("proximity", c(species_ids, gene_ids), {
+    cached("proximity", c(species_ids, gene_ids, use_positions), {
         # Prefilter distances by species and gene.
-        distances <- geposan::distances[
+        data <- geposan::distances[
             species %chin% preset$species_ids & gene %chin% preset$gene_ids
         ]
 
         # Compute the score as described above.
 
-        distances <- distances[, .(mean_distance = mean(distance)), by = "gene"]
+        data <- if (use_positions) {
-        max_distance <- distances[, max(mean_distance)]
+            data[, .(mean_distance = mean(position)), by = "gene"]
-        distances[, score := 1 - mean_distance / max_distance]
+        } else {
+            data[, .(mean_distance = mean(distance)), by = "gene"]
+        }
+
+        max_distance <- data[, max(mean_distance)]
+        data[, score := 1 - mean_distance / max_distance]
 
         if (!is.null(progress)) {
             # We do everything in one go, so it's not possible to report
@@ -25,6 +30,6 @@ proximity <- function(preset, progress = NULL) {
             progress(1.0)
         }
 
-        distances[, .(gene, score)]
+        data[, .(gene, score)]
     })
 }