filter ontology in barplot

galaxyproteomics · Apr 1, 2019 · 53f705d · 53f705d
1 parent f7399a0
commit 53f705d
Show file tree

Hide file tree

Showing 10 changed files with 203 additions and 24 deletions.
diff --git a/metaquantome/cli.py b/metaquantome/cli.py
@@ -244,6 +244,9 @@ def parse_args_cli():
                           ' heatmap and PCA plot, so the colors can be tweaked to match. ')
     bar.add_argument('--target_rank',
                      help="(Tax bar and FT dist). Taxonomic rank to restrict to in the plot. ")
+    bar.add_argument('--target_onto', choices=["mf", "bp", "cc"],
+                     help="(Function and FT dist bar only) " +
+                          "Ontology to restrict to, for function distribution.")
     bar.add_argument("--whichway", choices=["f_dist", "t_dist"],
                      help="(FT dist only) " +
                           "Which distribution - functional distribution for a taxon (f_dist) or " +
@@ -254,9 +257,6 @@ def parse_args_cli():
     bar.add_argument("--id",
                      help="(FT dist bar only) " +
                           "Taxonomic or functional term id - either a NCBI taxID or a GO term id (GO:XXXXXXX)")
-    bar.add_argument('--target_onto', choices=["mf", "bp", "cc"],
-                     help="(FT dist bar only) " +
-                          "Ontology to restrict to, for function distribution.")
 
     volc = parser_viz.add_argument_group('Volcano Plot')
     volc.add_argument('--fc_name',

diff --git a/metaquantome/data/test/eggnog_out.tab b/metaquantome/data/test/eggnog_out.tab
diff --git a/metaquantome/data/test/eggnog_viz_file.tab b/metaquantome/data/test/eggnog_viz_file.tab
@@ -0,0 +1,21 @@
+id	name	namespace	NS_mean	WS_mean	int737NS	int737WS	int852NS	int852WS	int867NS	int867WS	int737NS_n_peptide	int737WS_n_peptide	int852NS_n_peptide	int852WS_n_peptide	int867NS_n_peptide	int867WS_n_peptide	int737NS_n_samp_children	int737WS_n_samp_children	int852NS_n_samp_children	int852WS_n_samp_children	int867NS_n_samp_children	int867WS_n_samp_children
+GO:0008152	metabolic process	biological_process	9707895600.63897	34.4676527340745	32.7280807192344	33.639915914834	33.8500846328958	35.3680642871384	32.6033492340827	33.6566592988174	1837	1837	1837	1837	1837	1837	3	3	3	3	3	3
+GO:0009058	biosynthetic process	biological_process	3977879569.3907	33.2068587142223	31.5038426691202	32.3344633958136	32.5457060492145	34.154051958603	31.2863832807592	32.2790138218569	576	576	576	576	576	576	NA	NA	NA	NA	NA	NA
+GO:0006807	nitrogen compound metabolic process	biological_process	3333526855.75693	33.089118314848	31.3274682605914	32.2651614156765	32.2393712217206	33.9854328532233	31.0581331702586	32.2877493545852	763	763	763	763	763	763	NA	NA	NA	NA	NA	NA
+GO:0006412	translation	biological_process	2440026495.81359	32.4930473371394	30.8105936229541	31.6128211401251	31.846608199857	33.4455891939318	30.5527566639519	31.5536159837728	336	336	336	336	336	336	NA	NA	NA	NA	NA	NA
+GO:0006950	response to stress	biological_process	1052880100.76931	31.3981405492947	29.6095404092094	30.439676321524	30.5689283794754	32.393517103388	29.4759257423614	30.3742688501823	124	124	124	124	124	124	NA	NA	NA	NA	NA	NA
+GO:0042221	response to chemical	biological_process	965521061.018174	30.8296096443541	29.4808523042029	29.9278611878217	30.4629201331805	31.8156195431379	29.3138771113971	29.7828763814873	85	85	85	85	85	85	NA	NA	NA	NA	NA	NA
+GO:0044710	metabolic process	biological_process	928705245.346089	30.6072989115392	29.3036916765872	30.0343848802313	30.4615672319471	31.2433837819293	29.2644431642638	30.237026652397	225	225	225	225	225	225	3	3	3	3	3	3
+GO:0071840	cellular component organization or biogenesis	biological_process	833223817.553338	31.5724843198215	29.5524724554893	31.0024286599174	30.1336470253456	32.3466932720115	28.99316994804	30.8738545987975	357	357	357	357	357	357	NA	NA	NA	NA	NA	NA
+GO:0005975	carbohydrate metabolic process	biological_process	379644473.58704	30.2823742633555	27.4581777552516	28.9388737660123	29.2889137336267	31.4249235572303	28.1538245092415	28.9540435785897	76	76	76	76	76	76	NA	NA	NA	NA	NA	NA
+GO:0006091	generation of precursor metabolites and energy	biological_process	373375658.545852	27.5844108414961	26.7198540418534	27.2161871812721	29.5462583955409	27.603990048666	27.749246320597	27.8613248072876	30	30	30	30	30	30	NA	NA	NA	NA	NA	NA
+GO:0006629	lipid metabolic process	biological_process	346873604.466979	28.4302522117954	27.5680638918355	28.0217901314741	29.1700984804642	28.9561582409451	27.8238760468995	28.120421411102	37	37	37	37	37	37	NA	NA	NA	NA	NA	NA
+GO:0065003	protein-containing complex assembly	biological_process	223362430.986751	29.4067502060857	27.7185066355636	29.0195327862999	28.1835372517938	30.0898188984621	27.1053181932927	28.7398209542951	83	83	83	83	83	83	NA	NA	NA	NA	NA	NA
+GO:0044711	biosynthetic process	biological_process	220756695.513963	28.7739551105881	27.1093723370704	28.4003539740075	28.4104810597675	29.3934280879723	27.2597605688579	28.2424502003084	53	53	53	53	53	53	NA	NA	NA	NA	NA	NA
+GO:0044723	carbohydrate metabolic process	biological_process	182807836.125096	28.93667511454	26.448399628704	27.9008315674069	28.2396393635986	29.9843751125869	27.0599307314212	27.7686749761679	34	34	34	34	34	34	NA	NA	NA	NA	NA	NA
+GO:0019222	regulation of metabolic process	biological_process	173056856.95311	28.9389001060965	27.0494629499546	28.2721701007768	27.836284601413	29.6501969737214	27.0673584710354	28.4903755600971	69	69	69	69	69	69	NA	NA	NA	NA	NA	NA
+GO:0006810	transport	biological_process	140537615.614438	28.7939728267842	27.1234873806423	27.9615606778428	27.4144285386699	29.5408914880366	26.524314216561	28.3984348617235	88	88	88	88	88	88	NA	NA	NA	NA	NA	NA
+GO:0016032	viral process	biological_process	111058995.875521	30.4491645013834	26.3289762021563	29.2487087831025	27.1434558706649	31.5367058114516	26.5844879797392	29.263908857592	48	48	48	48	48	48	NA	NA	NA	NA	NA	NA
+GO:0016070	RNA metabolic process	biological_process	91888884.7293597	28.1369040778828	26.1677866053689	27.693875726079	26.8519757597805	28.5989127651248	26.2381122887607	27.9638332212402	81	81	81	81	81	81	NA	NA	NA	NA	NA	NA
+GO:0006461	protein-containing complex assembly	biological_process	88525220.719722	27.5427325704944	26.2738037562703	27.2967939297801	27.0298059946766	27.9798815285066	25.4990572646483	27.2258576018008	40	40	40	40	40	40	NA	NA	NA	NA	NA	NA
+GO:0006457	protein folding	biological_process	75079182.3365918	29.268835134676	25.7050718748645	28.2845189662675	26.7348727386628	30.2745828042056	25.8105457595204	28.2294977490008	32	32	32	32	32	32	NA	NA	NA	NA	NA	NA
diff --git a/metaquantome/data/test/test_eggnog_viz.png b/metaquantome/data/test/test_eggnog_viz.png
diff --git a/metaquantome/data/test/viz_f_filt.tab b/metaquantome/data/test/viz_f_filt.tab
@@ -0,0 +1,3 @@
+id	name	namespace	NS_mean	WS_mean	int737NS	int737WS	int852NS	int852WS	int867NS	int867WS	int737NS_n_peptide	int737WS_n_peptide	int852NS_n_peptide	int852WS_n_peptide	int867NS_n_peptide	int867WS_n_peptide	int737NS_n_samp_children	int737WS_n_samp_children	int852NS_n_samp_children	int852WS_n_samp_children	int867NS_n_samp_children	int867WS_n_samp_children
+GO:0003824	catalytic activity	molecular_function	1385407584.13192	31.4539840896763	29.8549354693473	30.8881079072179	31.0695559374115	32.1062472477587	29.7951982154145	31.044367271616	460	460	460	460	460	460	6	6	6	6	6	6
+GO:0003735	structural constituent of ribosome	molecular_function	1326838449.29867	31.821379782899	29.9285877846662	30.9700260743199	30.9954165750759	32.7495445116799	29.607449636552	30.9400897387871	223	223	223	223	223	223	NA	NA	NA	NA	NA	NA
diff --git a/metaquantome/modules/run_viz.py b/metaquantome/modules/run_viz.py
@@ -22,7 +22,7 @@ def run_viz(plottype, img, infile, strip=None,
     r_script_path = os.path.join(BASE_DIR, 'modules', 'viz.R')
     cmd = ['Rscript', '--vanilla', r_script_path, plottype, img, infile]
     if plottype == "bar":
-        cmd += [mode, meancol, nterms, width, height, target_rank, barcol, tabfile]
+        cmd += [mode, meancol, nterms, width, height, target_rank, target_onto, barcol, tabfile]
     elif plottype == "volcano":
         cmd += [str(textannot), fc_name, flip_fc, gosplit, width, height, tabfile]
     elif plottype == "heatmap":

diff --git a/metaquantome/modules/viz.R b/metaquantome/modules/viz.R
@@ -98,12 +98,31 @@ check_ranks <- function(df, target_rank) {
     }
 }
 
+short_onto_to_long <- c("bp" = "biological_process",
+                        "mf"="molecular_function",
+                        "cc"="cellular_component")
+
+filter_to_desired_onto <- function(df, target_onto) {
+    if (is.null(target_onto)) {
+        stop('must provide target ontology for function or function-taxonomy mode',
+             call. = FALSE)
+    }
+    long_onto <- short_onto_to_long[target_onto]
+    df <- df[df[, "namespace"] == long_onto, ]
+    # also, remove any bp, cc, or mf rows
+    df <- df[!(df[, "name"] %in% short_onto_to_long), ]
+    if (nrow(df) == 0) {
+        stop(paste0("No terms in the dataframe come from the desired ontology (", target_onto, ")"))
+    }
+    return(df)
+}
+
 ####### ==================== #######
 #              BARPLOT             #
 ####### ==================== #######
 mq_barplot <- function(df, img, mode, meancol,
-                       nterms, width, height, target_rank, int_barcol,
-                       tabfile){
+                       nterms, width, height, target_rank, target_onto,
+                       int_barcol, tabfile){
     if (!(meancol %in% names(df))){
         stop('Mean column name not found in dataframe. Check spelling and try again.',
              call. = FALSE)
@@ -114,8 +133,12 @@ mq_barplot <- function(df, img, mode, meancol,
         # filter
         df <- df[df[, "rank"] == target_rank, ]
     }
+    if (mode == "f") {
+        df <- filter_to_desired_onto(df, target_onto)
+    }
     # exponentiate
     df[, meancol] <- 2^df[, meancol]
+
     # reorder, for taking the top N terms
     reord <- df[order(df[, meancol], decreasing=TRUE), ]
     # take top N terms or number of rows, whichever is less
@@ -132,7 +155,7 @@ mq_barplot <- function(df, img, mode, meancol,
 
     ggplot(sub_reord) +
       geom_bar(aes_(x = reorder(sub_reord[, barnamecol], -sub_reord[, meancol]),
-                    y = as.name(meancol)), stat = "identity", fill = barcol, col = "black") +
+                    y = as.name(meancol)), stat = "identity", fill = barcol, col = "black", position = "dodge") +
       theme_bw() +
       labs(x = xlab, y = "Total Peptide Intensity") +
       theme(axis.text.x = element_text(angle = X_AXIS_ROT, hjust = 1))
@@ -166,14 +189,18 @@ barplot_cli <- function(args){
     width <- as.numeric(args[7])
     height <- as.numeric(args[8])
     target_rank <- args[9]
-    barcol <- as.numeric(args[10])
-    tabfile <- args[11]
+    if (target_rank == "None") target_rank <- NULL
+    target_onto <- args[10]
+    if (target_onto == "None") target_onto <- NULL
+    barcol <- as.numeric(args[11])
+    tabfile <- args[12]
     if (tabfile == "None") tabfile <- NULL
     plt <- mq_barplot(df, img=img, mode=mode,
                       meancol=meancol,
                       nterms=nterms,
                       height=height, width=width,
-                      target_rank=target_rank, int_barcol=barcol,
+                      target_rank=target_rank, target_onto = target_onto,
+                      int_barcol=barcol,
                       tabfile=tabfile)
 }
 
@@ -475,10 +502,6 @@ volcano_cli <- function(args){
 ####### ==================== #######
 #              FT DIST             #
 ####### ==================== #######
-short_onto_to_long <- c("bp" = "biological_process",
-                        "mf"="molecular_function",
-                        "cc"="cellular_component")
-
 mq_ft_dist <- function(df, img, whichway, name, id, meancol,
                        nterms, width, height, target_rank, target_onto,
                        int_barcol, tabfile){
@@ -515,13 +538,7 @@ mq_ft_dist <- function(df, img, whichway, name, id, meancol,
         check_ranks(df, target_rank)
         df <- df[df[, "rank"] == target_rank, ]
     } else if (whichway == "f_dist"){
-        long_onto <- short_onto_to_long[target_onto]
-        df <- df[df[, "namespace"] == long_onto, ]
-        # also, remove any bp, cc, or mf rows
-        df <- df[!(df[, "name"] %in% short_onto_to_long), ]
-        if (nrow(df) == 0) {
-            stop(paste0("No terms in the dataframe come from the desired ontology (", target_onto, ")"))
-        }
+        df <- filter_to_desired_onto(df, target_onto)
     } else {
     	stop("Wrong whichway - should be t_dist or f_dist.")
     }

diff --git a/tests/travis/testCLI.py b/tests/travis/testCLI.py
@@ -76,6 +76,22 @@ def testVizTabfile(self):
         self.assertEqual(b'3', nline.stdout.strip().split()[0])
         os.remove(tabfile)
 
+    def testFuncBar(self):
+        infile = testfile('eggnog_out.tab')
+        imgfile = testfile('test_eggnog_viz.png')
+        samps = testfile('rudney_samples.tab')
+        tabfile = testfile("eggnog_viz_file.tab")
+        cmd = ' '.join([
+            'python3 metaquantome/cli.py viz -m f --plottype bar '
+            '--infile', infile,
+            '--img', imgfile,
+            '--samps', samps,
+            '--nterms 20 --meancol NS_mean --target_onto bp',
+            '--tabfile', tabfile
+        ])
+        test_status = subprocess.call(cmd, shell=True)
+        self.assertEqual(test_status, 0)
+
     def testHeatmapViz(self):
         infile = testfile('ec_ttest_tested.tab')
         imgfile = testfile('cli_heatmap_viz.png')

diff --git a/tests/travis/testExpand.py b/tests/travis/testExpand.py
@@ -57,8 +57,8 @@ def testSingleInt(self):
         self.assertEqual(go_df.loc["GO:0008152"]['int'], np.log2(100))
 
     def testMultipleInt(self):
-        func=testfile('multiple_func.tab')
-        int=testfile('multiple_int.tab')
+        func = testfile('multiple_func.tab')
+        int = testfile('multiple_int.tab')
         go_df = expand.expand('f', sinfo='{"s1": ["int1", "int2", "int3"]}', int_file=int, pep_colname_int='peptide',
                               pep_colname_func='peptide', pep_colname_tax='peptide', data_dir=TEST_DIR, func_file=func,
                               func_colname='go', ontology='go')
@@ -84,10 +84,11 @@ def testNopep(self):
     def testSlimDown(self):
         func=testfile('func_eggnog.tab')
         int=testfile('int_eggnog.tab')
+        outfile=testfile('eggnog_out.tab')
         sinfo='{"NS": ["int737NS", "int852NS", "int867NS"], "WS": ["int737WS", "int852WS", "int867WS"]}'
         go_df = expand.expand('f', sinfo=sinfo, int_file=int, pep_colname_int='peptide', pep_colname_func='peptide',
                               pep_colname_tax='peptide', data_dir=TEST_DIR, func_file=func, func_colname='go', ontology='go',
-                              slim_down=True)
+                              slim_down=True, outfile=outfile)
         # test that all go terms are in slim
         # load slim
         returned_gos = set(go_df['id'])

diff --git a/tests/travis/testRunViz.py b/tests/travis/testRunViz.py
@@ -1,5 +1,6 @@
 import unittest
 import os
+import pandas as pd
 
 from metaquantome.util.testutils import testfile, TTEST_SINFO
 from metaquantome.modules.run_viz import run_viz
@@ -21,6 +22,33 @@ def testBasicTaxBar(self):
                 nterms='2', meancol='samp1_mean',
                 target_rank="genus", barcol="6")
 
+    def testFuncBar(self):
+        infile = testfile('eggnog_out.tab')
+        tabfile = testfile('viz_f_filt.tab')
+        # bp
+        run_viz('bar', self.img, infile, mode='f',
+                nterms='2', meancol='NS_mean',
+                target_onto='bp', barcol='1', tabfile=tabfile)
+        df = pd.read_csv(tabfile, sep='\t')
+        namespace = df.namespace.unique()
+        self.assertEqual(namespace, ['biological_process'])
+
+        # cc
+        run_viz('bar', self.img, infile, mode='f',
+                nterms='2', meancol='NS_mean',
+                target_onto='cc', barcol='1', tabfile=tabfile)
+        df = pd.read_csv(tabfile, sep='\t')
+        namespace = df.namespace.unique()
+        self.assertEqual(namespace, ['cellular_component'])
+
+        # mf
+        run_viz('bar', self.img, infile, mode='f',
+                nterms='2', meancol='NS_mean',
+                target_onto='mf', barcol='1', tabfile=tabfile)
+        df = pd.read_csv(tabfile, sep='\t')
+        namespace = df.namespace.unique()
+        self.assertEqual(namespace, ['molecular_function'])
+
     def testVolcano(self):
         infile = testfile('cli_mult_test_out.tab')
         run_viz('volcano', self.img, infile,
@@ -70,6 +98,13 @@ def testFtDist(self):
                 target_rank="genus",
                 nterms="all")
 
+        # f dist
+        run_viz('ft_dist', self.img, infile,
+                meancol="s1_mean",
+                whichway='f_dist',
+                target_onto="bp",
+                nterms="all", id=209)
+
         # test tabfile
         tabfile = testfile("tmp")
         run_viz('ft_dist', self.img, infile,