Improve ez_histograms (#662)

* simplify rownames management * adding test for misleading column ids (duplicates) * switch to psych statistics * update all tests * Update ez_histograms.xml * Update ez_histograms.xml
ARTbio · Feb 8, 2024 · 5e25392 · 5e25392
1 parent 443759a
commit 5e25392
Show file tree

Hide file tree

Showing 17 changed files with 1,176 additions and 154 deletions.
diff --git a/tools/ez_histograms/ez_histograms.R b/tools/ez_histograms/ez_histograms.R
@@ -2,7 +2,7 @@ library(ggplot2)
 library(reshape2)
 library(dplyr)
 library(scales)
-library(vtable)
+library(psych)
 library(optparse)
 
 options(show.error.messages = FALSE,
@@ -116,17 +116,9 @@ test_header <- function(file) {
   }
 }
 
-test_rownames <- function(file) {
-  data <- read.delim(file = file, header = FALSE, row.names = NULL, nrows = 2)
-  if (is.na(as.numeric(data[2, 1]))) {
-    return(1)
-  } else {
-    return(NULL)
-  }
-}
-
 ##### prepare input data
-data <- read.delim(file = opt$file, header = test_header(opt$file), row.names = test_rownames(opt$file))
+
+data <- read.delim(file = opt$file, header = test_header(opt$file))
 data <- data %>% select(where(is.numeric))  # remove non numeric columns
 mdata <- melt(data)
 
@@ -162,8 +154,10 @@ pdf(opt$pdf, width = width, height = height)
 print(p + facet_wrap(~variable, ncol = ncol, scales = "free"))
 dev.off()
 
-# Summary statistics with vtable package
-summary_df <- sumtable(data, digits = 8, out = "return", add.median = TRUE,
-                       summ.names = c("N", "Mean", "Std. Dev.", "Min", "Pctl. 25",
-                                      "Median", "Pctl. 75", "Max"))
+# Summary statistics with psych package
+summary_df <- describe(x = data, skew = FALSE, ranges = FALSE, quant = c(.25, .50, .75))
+summary_df <- cbind(var_names = rownames(summary_df), summary_df)
+colnames(summary_df)[2] <- "var_num"
+summary_df <- summary_df[, -6]
+summary_df[, 4:8] <- format(summary_df[, 4:8], scientific = TRUE)
 write.table(summary_df, file = opt$summary, sep = "\t", quote = FALSE, row.names = FALSE)
diff --git a/tools/ez_histograms/ez_histograms.xml b/tools/ez_histograms/ez_histograms.xml
@@ -1,15 +1,15 @@
 <tool id="ez_histograms" name="ez_histograms" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="@PROFILE@">
     <macros>
         <token name="@TOOL_VERSION@">3.4.4</token>
-        <token name="@VERSION_SUFFIX@">0</token>
+        <token name="@VERSION_SUFFIX@">1</token>
         <token name="@PROFILE@">23.0</token>
     </macros>
     <requirements>
         <requirement type="package" version="3.4.4">r-ggplot2</requirement>
         <requirement type="package" version="1.4.4">r-reshape2</requirement>
         <requirement type="package" version="1.1.4">r-dplyr</requirement>
         <requirement type="package" version="1.3.0">r-scales</requirement>
-        <requirement type="package" version="1.4.6">r-vtable</requirement>
+        <requirement type="package" version="2.4.1">r-psych</requirement>
         <requirement type="package" version="1.7.4">r-optparse</requirement>
     </requirements>
 
@@ -108,6 +108,13 @@
             <output name="pdf" file="single_headed_col.pdf"/>
             <output name="summary" file="summary_6.tsv"/>
        </test>
+        <test expect_num_outputs="2">
+            <param name="file" value="misleading_ids.tsv"/>
+            <param name="plot_options_selector" value="density"/>
+            <param name="xscale" value="cartesian"/>
+            <output name="pdf" file="misleading_ids.pdf"/>
+            <output name="summary" file="summary_7.tsv"/>
+       </test>
     </tests>
     <help><![CDATA[
 **What it does**
@@ -145,7 +152,7 @@ scales
 The ez_histograms Galaxy tool returns
 
 - A pdf file with plots faceted on three columns
-- A tsv file with statistics summary of the variables generated by the R package vtables
+- A tsv file with statistics summary of the variables generated by the R package psych
 
     ]]></help>
     <citations>

diff --git a/tools/ez_histograms/test-data/classic.pdf b/tools/ez_histograms/test-data/classic.pdf
diff --git a/tools/ez_histograms/test-data/count.pdf b/tools/ez_histograms/test-data/count.pdf
diff --git a/tools/ez_histograms/test-data/headless.pdf b/tools/ez_histograms/test-data/headless.pdf
diff --git a/tools/ez_histograms/test-data/large.pdf b/tools/ez_histograms/test-data/large.pdf
diff --git a/tools/ez_histograms/test-data/misleading_ids.pdf b/tools/ez_histograms/test-data/misleading_ids.pdf
diff --git a/tools/ez_histograms/test-data/misleading_ids.tsv b/tools/ez_histograms/test-data/misleading_ids.tsv
diff --git a/tools/ez_histograms/test-data/rowheadless.pdf b/tools/ez_histograms/test-data/rowheadless.pdf
diff --git a/tools/ez_histograms/test-data/single_headed_col.pdf b/tools/ez_histograms/test-data/single_headed_col.pdf
diff --git a/tools/ez_histograms/test-data/summary_1.tsv b/tools/ez_histograms/test-data/summary_1.tsv
@@ -1,7 +1,7 @@
-Variable	N	Mean	Std. Dev.	Min	Pctl. 25	Median	Pctl. 75	Max
-BaseMean	990	1091.5132	2625.4763	0	31.199603	249.39264	1054.0101	33915.214
-log2FC	849	-0.38999858	1.9953097	-5.2835189	-2.0088992	-0.86544745	1.2737699	5.6079286
-StdErr	849	0.30667747	0.1285766	0.089282127	0.1962894	0.28928034	0.39630761	0.56482292
-Wald.Stats	849	-0.95820431	6.3140778	-21.603869	-5.9077253	-4.475739	5.2726083	19.585597
-P.value	848	0.030010804	0.15185376	0.00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000016518389	0.0000000000036649168	0.000000012325875	0.0000022785953	0.98727107
-P.adj	813	0.00004223522	0.000078050135	0.0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000034761298	0.000000000255376	0.00000044934286	0.000034873651	0.00028949939
+var_names	var_num	n	mean	sd	Q0.25	Q0.5	Q0.75
+BaseMean	1	990	 1.091513e+03	2.625476e+03	 3.119960e+01	 2.493926e+02	1.054010e+03
+log2FC	2	849	-3.899986e-01	1.995310e+00	-2.008899e+00	-8.654475e-01	1.273770e+00
+StdErr	3	849	 3.066775e-01	1.285766e-01	 1.962894e-01	 2.892803e-01	3.963076e-01
+Wald.Stats	4	849	-9.582043e-01	6.314078e+00	-5.907725e+00	-4.475739e+00	5.272608e+00
+P.value	5	848	 3.001080e-02	1.518538e-01	 3.664917e-12	 1.232588e-08	2.278595e-06
+P.adj	6	813	 4.223522e-05	7.805013e-05	 2.553760e-10	 4.493429e-07	3.487365e-05
diff --git a/tools/ez_histograms/test-data/summary_2.tsv b/tools/ez_histograms/test-data/summary_2.tsv
@@ -1,16 +1,16 @@
-Variable	N	Mean	Std. Dev.	Min	Pctl. 25	Median	Pctl. 75	Max
-GCB_Mg_S2	1000	672.224	1727.7831	0	69	236.5	700	38072
-GCB_Mg_S7	1000	830.225	1848.5916	0	91.5	317	857.25	34261
-GCB_Mg_S12	1000	685.564	1662.1423	0	71	253	686.25	33558
-GCB_S1	1000	450.654	1224.5977	0	38	152	459.5	29464
-GCB_S6	1000	547.793	1344.4334	0	51	194	550	28567
-GCB_S11	1000	481.844	1310.3546	0	40	152.5	462.5	29614
-GW_S5	1000	698.673	1534.1178	0	63.75	237.5	697.75	26119
-GW_S10	1000	573.137	1477.6949	0	49.75	176.5	504	29920
-GW_S15	1000	718.612	1730.7819	0	69	233.5	676	33878
-NS_S4	1000	824.02	1624.4871	0	100	336	914	28284
-NS_S9	1000	704.673	1266.9907	0	93.5	305	796.25	21678
-NS_S14	1000	883.989	1636.9581	0	114.75	366.5	1016	25492
-Starch_S3	1000	905.783	1464.8208	1	135.75	430	1099	22970
-Starch_S8	1000	973.323	1680.7842	0	138.75	426.5	1092.75	24999
-Starch_S13	1000	752.171	1404.0569	0	93	310.5	846.25	22556
+var_names	var_num	n	mean	sd	Q0.25	Q0.5	Q0.75
+GCB_Mg_S2	1	1000	6.72224e+02	1.727783e+03	6.9000e+01	2.365e+02	7.00000e+02
+GCB_Mg_S7	2	1000	8.30225e+02	1.848592e+03	9.1500e+01	3.170e+02	8.57250e+02
+GCB_Mg_S12	3	1000	6.85564e+02	1.662142e+03	7.1000e+01	2.530e+02	6.86250e+02
+GCB_S1	4	1000	4.50654e+02	1.224598e+03	3.8000e+01	1.520e+02	4.59500e+02
+GCB_S6	5	1000	5.47793e+02	1.344433e+03	5.1000e+01	1.940e+02	5.50000e+02
+GCB_S11	6	1000	4.81844e+02	1.310355e+03	4.0000e+01	1.525e+02	4.62500e+02
+GW_S5	7	1000	6.98673e+02	1.534118e+03	6.3750e+01	2.375e+02	6.97750e+02
+GW_S10	8	1000	5.73137e+02	1.477695e+03	4.9750e+01	1.765e+02	5.04000e+02
+GW_S15	9	1000	7.18612e+02	1.730782e+03	6.9000e+01	2.335e+02	6.76000e+02
+NS_S4	10	1000	8.24020e+02	1.624487e+03	1.0000e+02	3.360e+02	9.14000e+02
+NS_S9	11	1000	7.04673e+02	1.266991e+03	9.3500e+01	3.050e+02	7.96250e+02
+NS_S14	12	1000	8.83989e+02	1.636958e+03	1.1475e+02	3.665e+02	1.01600e+03
+Starch_S3	13	1000	9.05783e+02	1.464821e+03	1.3575e+02	4.300e+02	1.09900e+03
+Starch_S8	14	1000	9.73323e+02	1.680784e+03	1.3875e+02	4.265e+02	1.09275e+03
+Starch_S13	15	1000	7.52171e+02	1.404057e+03	9.3000e+01	3.105e+02	8.46250e+02
diff --git a/tools/ez_histograms/test-data/summary_3.tsv b/tools/ez_histograms/test-data/summary_3.tsv
@@ -1,7 +1,7 @@
-Variable	N	Mean	Std. Dev.	Min	Pctl. 25	Median	Pctl. 75	Max
-V2	1000	1286.6999	2744.7392	6.5122425	100.50742	409.92217	1309.8068	33915.214
-V3	1000	-0.4092237	1.9730612	-5.2835189	-2.0884579	-0.946016	1.2600548	5.6079286
-V4	1000	0.30466439	0.1311528	0.089282127	0.19205135	0.28205562	0.40370597	0.56482292
-V5	1000	-0.9419943	6.2085133	-21.603869	-5.7808572	-4.5659907	5.0667272	19.585597
-V6	1000	0.0000018003049	0.000003260254	0.00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000016518389	0.00000000002981881	0.000000042451763	0.0000019035803	0.000013756861
-V7	1000	0.000042996682	0.00007258655	0.0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000034761298	0.0000000025022835	0.000001784909	0.000053393931	0.00028949939
+var_names	var_num	n	mean	sd	Q0.25	Q0.5	Q0.75
+V2	1	1000	 1.286700e+03	2.744739e+03	 1.005074e+02	 4.099222e+02	1.309807e+03
+V3	2	1000	-4.092237e-01	1.973061e+00	-2.088458e+00	-9.460160e-01	1.260055e+00
+V4	3	1000	 3.046644e-01	1.311528e-01	 1.920513e-01	 2.820556e-01	4.037060e-01
+V5	4	1000	-9.419943e-01	6.208513e+00	-5.780857e+00	-4.565991e+00	5.066727e+00
+V6	5	1000	 1.800305e-06	3.260254e-06	 2.981881e-11	 4.245176e-08	1.903580e-06
+V7	6	1000	 4.299668e-05	7.258655e-05	 2.502284e-09	 1.784909e-06	5.339393e-05