Merge pull request #278 from OHDSI/issue_266

modified description, readme, and mockOmopSketch

DESCRIPTION

@@ -54,7 +54,7 @@ Suggests:
     omock (>= 0.3.0),
     covr,
     ggplot2,
-    visOmopResults (>= 0.4.0)
+    visOmopResults (>= 0.5.0)
 Config/testthat/edition: 3
 Config/testthat/parallel: true
 Imports: 
@@ -65,7 +65,7 @@ Imports:
     dplyr,
     glue,
     lifecycle,
-    omopgenerics (>= 0.3.1),
+    omopgenerics (>= 0.4.1),
     PatientProfiles (>= 1.2.1),
     purrr,
     rlang,
@@ -77,5 +77,4 @@ Depends:
 URL: https://OHDSI.github.io/OmopSketch/
 BugReports: https://github.com/OHDSI/OmopSketch/issues
 VignetteBuilder: knitr
-Remotes: 
-    darwin-eu/omopgenerics
+

R/mockOmopSketch.R

@@ -48,10 +48,48 @@ mockOmopSketch <- function(con = NULL,
     omock::mockProcedureOccurrence(seed = seed) |>
     omock::mockVisitOccurrence(seed = seed) |>
     # Create device exposure table - empty (Eunomia also has it empty)
-    omopgenerics::emptyOmopTable("device_exposure")
+    omopgenerics::emptyOmopTable("device_exposure")|>
+    checkColumns()
+
 
   # WHEN WE SUPORT LOCAL CDMs WE WILL HAVE TO ACCOUNT FOR THAT HERE
   cdm <- CDMConnector::copy_cdm_to(con = con, cdm = cdm, schema = writeSchema)
 
   return(cdm)
 }
+
+checkColumns <- function(cdm_local){
+  info <- omopgenerics::omopTableFields() |>
+    dplyr::filter(.data$type == "cdm_table") |>
+    dplyr::mutate(cdm_datatype = dplyr::case_when(
+      .data$cdm_datatype == "integer" ~ "NA_integer_",
+      grepl("varchar", .data$cdm_datatype) ~ "NA_character_",
+      .default = "NA"
+    ))
+  for (table in names(cdm_local)){
+    cols <-  info |>
+      dplyr::filter(.data$cdm_table_name == table)|>
+      dplyr::select(cdm_field_name,cdm_datatype)
+
+    missing_cols <- cols|>
+      dplyr::filter(!(cdm_field_name %in% colnames(cdm_local[[table]])))
+
+    if (nrow(missing_cols) > 0) {
+
+      missing_tbl <- tibble::tibble(
+        !!!rlang::set_names(
+          lapply(missing_cols$cdm_datatype, function(datatype) {
+            eval(parse(text = datatype))
+          }),
+          missing_cols$cdm_field_name
+        )
+      )
+
+      cdm_local[[table]] <- dplyr::bind_cols(cdm_local[[table]], missing_tbl)
+
+    }
+  }
+  return(cdm_local)
+}
+
+

R/plotConceptSetCounts.R

@@ -54,13 +54,13 @@ plotConceptSetCounts <- function(result,
     ))
   }
 
-  result1 <- result |> omopgenerics::splitAdditional()
+  result1 <- result |> omopgenerics::splitAll()
   # Detect if there are several time intervals
   if("time_interval" %in% colnames(result1)){
     # Line plot where each concept is a different line
     p <- result1 |>
       dplyr::filter(.data$time_interval != "overall") |>
-      omopgenerics::uniteAdditional(cols = c("time_interval", "standard_concept_name", "standard_concept_id", "source_concept_name", "source_concept_id", "domain_id")) |>
+      omopgenerics::pivotEstimates() |>
       visOmopResults::scatterPlot(x = "time_interval",
                                   y = "count",
                                   line   = TRUE,
@@ -71,17 +71,28 @@ plotConceptSetCounts <- function(result,
                                   colour = colour)
   }else{
     if("standard_concept_name" %in% colnames(result1)){
-      p <- result |>
+      p <- result1 |>
+        omopgenerics::pivotEstimates() |>
         visOmopResults::barPlot(x = c("standard_concept_name", "standard_concept_id"),
                                 y = "count",
                                 facet = facet,
                                 colour = colour)
+      p$data <- p$data |>
+        dplyr::mutate(
+          standard_concept_name_standard_concept_id = factor(
+            .data$standard_concept_name_standard_concept_id,
+            levels = c("overall - overall", sort(setdiff(.data$standard_concept_name_standard_concept_id, "overall - overall")))
+          )
+        )
+
     }else{
-      p <- result |>
+      p <- result1 |>
         visOmopResults::barPlot(x = "codelist_name",
                                 y = "count",
                                 facet = facet,
                                 colour = colour)
+      p$data <- p$data |>
+       dplyr::arrange(.data$codelist_name)
     }
     p <- p +
       ggplot2::labs(

R/plotInObservation.R

@@ -59,8 +59,7 @@ plotInObservation <- function(result,
 
   # plot
   if(length(unique(result$additional_level)) > 1 ){
-    result |>
-      dplyr::mutate(additional_level = as.character(gsub("-01$","",as.Date(gsub(" to.*","",.data$additional_level))))) |>
+   p <- result |>
       dplyr::filter(.data$estimate_name == "count") |>
       visOmopResults::scatterPlot(
         x = "time_interval",
@@ -78,6 +77,21 @@ plotInObservation <- function(result,
         y = variable,
         x = "Date"
       )
+   p$data <- p$data |>
+     dplyr::arrange(.data$time_interval) |>
+     dplyr::mutate(
+       show_label = seq_along(.data$time_interval) %% ceiling(nrow(p$data) / 20) == 0
+     )
+
+   p <- p +
+     ggplot2::scale_x_discrete(
+       breaks = p$data$time_interval[p$data$show_label]
+     ) +
+     ggplot2::theme(
+       axis.text.x = ggplot2::element_text(angle = 90, hjust = 1, size = 8),
+       plot.margin = ggplot2::margin(t = 5, r = 5, b = 30, l = 5)
+     )
+   p
   }else{
     result |>
       dplyr::filter(.data$estimate_name == "count") |>

R/plotRecordCount.R

@@ -59,6 +59,30 @@ plotRecordCount <- function(result,
         y = "Number records",
         x = "Date"
       )
+    p$data <- p$data |>
+      dplyr::arrange(.data$time_interval) |>
+      dplyr::group_by(.data$omop_table) |>
+      dplyr::mutate(
+        show_label = if (dplyr::cur_group_id() == 1) {
+          seq_along(.data$time_interval) %% ceiling(dplyr::n() / 20) == 0
+        } else {
+          FALSE
+        }
+      ) |>
+      dplyr::ungroup()
+
+    # Modify the plot
+    p <- p +
+      ggplot2::scale_x_discrete(
+        breaks = p$data$time_interval[p$data$show_label],
+        labels = p$data$time_interval[p$data$show_label]
+      ) +
+      ggplot2::theme(
+        axis.text.x = ggplot2::element_text(angle = 90, hjust = 1, size = 8),
+        plot.margin = ggplot2::margin(t = 5, r = 5, b = 30, l = 5)
+      )
+
+
   }else{
     p <- result |>
       visOmopResults::barPlot(x = "variable_name",
@@ -70,5 +94,7 @@ plotRecordCount <- function(result,
         x = ""
       )
   }
+
   p
 }
+

README.Rmd

@@ -16,13 +16,10 @@ knitr::opts_chunk$set(
 # OmopSketch <a href="https://OHDSI.github.io/OmopSketch/"><img src="man/figures/logo.png" align="right" height="138" alt="OmopSketch website" /></a>
 
 <!-- badges: start -->
-[![Lifecycle: experimental](https://img.shields.io/badge/lifecycle-experimental-orange.svg)](https://lifecycle.r-lib.org/articles/stages.html#experimental)
-[![R-CMD-check](https://github.com/OHDSI/OmopSketch/actions/workflows/R-CMD-check.yaml/badge.svg)](https://github.com/OHDSI/OmopSketch/actions/workflows/R-CMD-check.yaml)
-[![CRAN status](https://www.r-pkg.org/badges/version/OmopSketch)](https://CRAN.R-project.org/package=OmopSketch)
-[![Codecov test coverage](https://codecov.io/gh/OHDSI/OmopSketch/branch/main/graph/badge.svg)](https://app.codecov.io/gh/OHDSI/OmopSketch?branch=main)
-<!-- badges: end -->
 
-### WARNING: this package is under-development and has only been tested using mock data
+[![Lifecycle: experimental](https://img.shields.io/badge/lifecycle-experimental-orange.svg)](https://lifecycle.r-lib.org/articles/stages.html#experimental) [![R-CMD-check](https://github.com/OHDSI/OmopSketch/actions/workflows/R-CMD-check.yaml/badge.svg)](https://github.com/OHDSI/OmopSketch/actions/workflows/R-CMD-check.yaml) [![CRAN status](https://www.r-pkg.org/badges/version/OmopSketch)](https://CRAN.R-project.org/package=OmopSketch) [![Codecov test coverage](https://codecov.io/gh/OHDSI/OmopSketch/branch/main/graph/badge.svg)](https://app.codecov.io/gh/OHDSI/OmopSketch?branch=main)
+
+<!-- badges: end -->
 
 The goal of OmopSketch is to characterise and visualise an OMOP CDM instance to asses if it meets the necessary criteria to answer a specific clinical question and conduct a certain study.
 
@@ -48,16 +45,19 @@ con <- dbConnect(duckdb(), eunomia_dir())
 cdm <- cdmFromCon(con = con, cdmSchema = "main", writeSchema = "main")
 cdm
 ```
+
 ### Snapshot
+
 We first create a snapshot of our database. This will allow us to track when the analysis has been conducted and capture details about the CDM version or the data release.
+
 ```{r}
 summariseOmopSnapshot(cdm) |>
   tableOmopSnapshot(type = "flextable")
 ```
 
-
 ### Characterise the clinical tables
-Once we have collected the snapshot information, we can start characterising the clinical tables of the CDM. By using `summariseClinicalRecords()` and `tableClinicalRecords()`, we can easily visualise the main characteristics of specific clinical tables. 
+
+Once we have collected the snapshot information, we can start characterising the clinical tables of the CDM. By using `summariseClinicalRecords()` and `tableClinicalRecords()`, we can easily visualise the main characteristics of specific clinical tables.
 
 ```{r}
 summariseClinicalRecords(cdm, c("condition_occurrence", "drug_exposure")) |>
@@ -67,45 +67,45 @@ summariseClinicalRecords(cdm, c("condition_occurrence", "drug_exposure")) |>
 We can also explore trends in the clinical table records over time.
 
 ```{r}
-summariseRecordCount(cdm, c("condition_occurrence", "drug_exposure")) |>
-  plotRecordCount(facet = "omop_table")
+summariseRecordCount(cdm, c("condition_occurrence", "drug_exposure"), interval = "years") |>
+  plotRecordCount(facet = "omop_table", colour = "cdm_name")
 ```
+
 ### Characterise the observation period
+
 After visualising the main characteristics of our clinical tables, we can explore the observation period details. OmopSketch provides several functions to have an overview the dataset study period.
 
 Using `summariseInObservation()` and `plotInObservation()`, we can gather information on the number of records per year.
 
 ```{r}
-summariseInObservation(cdm$observation_period, output = "records") |>
-  plotInObservation()
+summariseInObservation(cdm$observation_period, output = "records", interval = "years") |>
+  plotInObservation(colour = "cdm_name")
 ```
-You can also visualise and explore the characteristics of the observation period per each individual in the database using `summariseObservationPeriod()`. 
+
+You can also visualise and explore the characteristics of the observation period per each individual in the database using `summariseObservationPeriod()`.
+
 ```{r}
 summariseObservationPeriod(cdm$observation_period) |>
   tableObservationPeriod(type = "flextable")
 ```
 
 Or if visualisation is preferred, you can easily build a histogram to explore how many participants have more than one observation period.
+
 ```{r}
 summariseObservationPeriod(cdm$observation_period) |>
-  plotObservationPeriod()
+  plotObservationPeriod(colour = "observation_period_ordinal")
 ```
 
 ### Characterise the concepts
+
 OmopSketch also provides functions to explore some of (or all) the concepts in the dataset.
+
 ```{r}
 acetaminophen <- c(1125315,  1127433, 1127078)
 
 summariseConceptSetCounts(cdm, conceptSet = list("acetaminophen" = acetaminophen)) |>
   filter(variable_name == "Number records") |> 
-  plotConceptSetCounts()
+  plotConceptSetCounts(colour = "codelist_name")
 ```
 
-### Characterise the population
-Finally, OmopSketch can also help us to characterise the population at the start and end of the observation period.
-```{r}
-summarisePopulationCharacteristics(cdm) |>
-  tablePopulationCharacteristics(type = "flextable")
-```
 As seen, OmopSketch offers multiple functionalities to provide a general overview of a database. Additionally, it includes more tools and arguments that allow for deeper exploration, helping to assess the database's suitability for specific research studies. For further information, please refer to the vignettes.
-