Merge pull request #1 from hadexversum/revs

Revs

NAMESPACE

@@ -19,6 +19,7 @@ export(fit_3_exp)
 export(fix_class_result)
 export(get_example_control)
 export(get_example_fit_k_params)
+export(get_example_fit_k_params_2)
 export(get_fit_results)
 export(get_fit_values_info)
 export(get_params_summary_image)

R/calculate_hires.R

@@ -40,6 +40,7 @@ calculate_hires <- function(fit_values,
       }
 
       tmp_params <- tmp_params %>%
+        filter(!class_name %in% c("invalid", "invalid_uc")) %>%
         arrange(nchar(sequence), class_name) %>%
         .[1, ]
 
@@ -119,13 +120,13 @@ calculate_hires <- function(fit_values,
           res <- data.frame(Protein = Protein,
                      State = State,
                      position = pos,
-                     n_1 = NA,
-                     k_1 = NA,
-                     n_2 = NA,
-                     k_2 = NA,
-                     n_3 = NA,
-                     k_3 = NA,
-                     k_est = NA,
+                     n_1 = class_example[["n_1"]],
+                     k_1 = class_example[["k_1"]],
+                     n_2 = class_example[["n_2"]],
+                     k_2 = class_example[["k_3"]],
+                     n_3 = class_example[["n_3"]],
+                     k_3 = class_example[["k_3"]],
+                     k_est = class_example[["k_est"]],
                      class_name = class_example[["class_name"]],
                      color = class_example[["color"]])
 

R/create_fit_dataset.R

@@ -28,6 +28,7 @@ create_fit_dataset <- function(kin_dat,
                                control = list(maxiter = 1000,  scale = "levenberg"),
                                trace = FALSE,
                                fractional = FALSE,
+                               omit_t_100 = FALSE,
                                workflow = 321){
 
   peptide_list <- kin_dat %>%
@@ -47,6 +48,7 @@ create_fit_dataset <- function(kin_dat,
                     trace = trace,
                     workflow = workflow,
                     fractional = fractional,
+                    omit_t_100 = omit_t_100,
                     edge_times = edge_times)
   }) %>% bind_rows() %>% mutate(id = 1:nrow(.)) %>% remove_rownames(.) %>%
     select(id, everything())

R/get_fit_results.R

@@ -5,6 +5,7 @@ get_fit_results <- function(fit_dat,
                             trace = FALSE,
                             workflow = 31,
                             fractional = TRUE,
+                            omit_t_100 = FALSE,
                             edge_times = c(min(fit_dat[["Exposure"]]), max(fit_dat[["Exposure"]]))){
 
   workflow <- match.arg(as.character(workflow), choices = c(31, 21, 321))
@@ -27,6 +28,11 @@ get_fit_results <- function(fit_dat,
                             fit_k_params))
   }
 
+  if(omit_t_100){
+    time_100 <- attr(fit_dat, "time_100")
+    fit_dat <- filter(fit_dat, Exposure < time_100)
+  }
+
   if(workflow == 31){
 
     fit_3 <- fit_3_exp(fit_dat,
@@ -145,7 +151,7 @@ fix_class_result <- function(fit_dat,
     color = "#000000"
   }
 
-  if(class_name == "invalid"){
+  if(class_name %in% c("invalid", "invalid_uc")){
     k_1 = n_1 = k_2 = n_2 = k_3 = n_3 = NA
     color = "#808080"
   }
@@ -200,6 +206,9 @@ detect_class <-  function(fit_dat, edge_times = NULL){
   if(du_100/ max_uptake < threshold & du_100 < 1) return("none")
   # if((du_100 - du_1)/max_uptake < threshold &  du_1 / du_100 > 1 - threshold) return("immediate")
 
+  accepted_fluctuation = 0.5 ## 0.5 Da doesnt significate exchange
+  if(du_100 < du_1 & !all(fit_dat[["deut_uptake"]] - du_100 < 0.5)) return("invalid_uc")
+
   class_name
 
 }

R/getters.R

@@ -11,6 +11,19 @@ get_example_fit_k_params <- function(){
 
 }
 
+#' Example k parameters
+#'
+#' @export
+
+get_example_fit_k_params_2 <- function(){
+
+  return(data.frame(
+    start = c(k_1 = 2, k_2 = 0.2, k_3 = 0.02),
+    lower = c(k_1 = 1, k_2 = 0.1, k_3 = 0.0001),
+    upper = c(k_1 = 30, k_2 = 1, k_3 = 0.1)))
+
+}
+
 #' Example control
 #'
 #' @export
@@ -76,7 +89,7 @@ get_3_n_params <- function(MaxUptake = NA){
 
     data.frame(
       start = c(n_1 = MaxUptake/3, n_2 = MaxUptake/3, n_3 = MaxUptake/3),
-      lower = c(n_1 = 0, n_2 = 0, n_3 = 0),
+      lower = c(n_1 = 1, n_2 = 1, n_3 = 1),
       upper = c(n_1 = MaxUptake, n_2 = MaxUptake, n_3 = MaxUptake)
     )
 
@@ -97,7 +110,7 @@ get_2_n_params <- function(MaxUptake = NA){
   } else {
     data.frame(
       start = c(n_1 = MaxUptake/2, n_2 = MaxUptake/2),
-      lower = c(n_1 = 0, n_2 = 0),
+      lower = c(n_1 = 1, n_2 = 1),
       upper = c(n_1 = MaxUptake, n_2 = MaxUptake)
     )
   }
@@ -116,7 +129,7 @@ get_1_n_params <- function(MaxUptake = NA){
   } else {
     data.frame(
       start = c(n_1 = MaxUptake*0.7),
-      lower = c(n_1 = 0),
+      lower = c(n_1 = 1),
       upper = c(n_1 = MaxUptake)
     )
   }

R/recreate_uc.R

@@ -8,6 +8,16 @@
 #' @description This function recovers fractional deuterium uptake from high-resolution level
 #' back to the peptide level. First, the values are aggregated into high-resolution
 #' level using selected aggregation method.
+#' If there are no calculated recovered values, probably the peptide was calssified as
+#' invalid - check that in fit_values_all param to be sure.
+#'
+#' @examples
+#' kin_dat <- prepare_kin_dat(alpha_dat)
+#' fit_values_all <- create_fit_dataset(kin_dat, control = get_example_control(),
+#'                                      fit_k_params = get_example_fit_k_params_2(),
+#'                                      fractional = TRUE)
+#' fit_dat <- kin_dat[kin_dat[["ID"]] == 54, ]
+#' calculate_uc_from_hires_peptide(fit_dat, fit_values_all, hires_method = "weighted")
 #'
 #' @seealso
 #' \code{\link{create_uc_from_hires_dataset}}
@@ -59,7 +69,6 @@ calculate_uc_from_hires_peptide <- function(fit_dat, ## uc filtered dat
              hr_diff = deut_uptake - hr_deut_uptake)
   }
 
-
   return(res)
 }