NJ tree tool

galaxy/data_managers/data_manager_conf.xml

@@ -1,6 +1,6 @@
 <?xml version="1.0"?>
 <data_managers>
-  <data_manager tool_file="data_manager_mentalist_build_db/data_manager/mentalist_build_db.xml" id="mentalist_build_db" version="0.1-SNAPSHOT">
+  <data_manager tool_file="data_manager_mentalist_build_db/data_manager/mentalist_build_db.xml" id="mentalist_build_db" version="0.1.8">
     <data_table name="mentalist_databases">
       <output>
         <column name="value" />
@@ -15,7 +15,7 @@
       </output>
     </data_table>
   </data_manager>
-  <data_manager tool_file="data_manager_mentalist_download_cgmlst/data_manager/mentalist_download_cgmlst.xml" id="mentalist_download_cgmlst" version="0.1-SNAPSHOT">
+  <data_manager tool_file="data_manager_mentalist_download_cgmlst/data_manager/mentalist_download_cgmlst.xml" id="mentalist_download_cgmlst" version="0.1.8">
     <data_table name="mentalist_databases">
       <output>
         <column name="value" />
@@ -30,7 +30,7 @@
       </output>
     </data_table>
   </data_manager>
-  <data_manager tool_file="data_manager_mentalist_download_pubmlst/data_manager/mentalist_download_pubmlst.xml" id="mentalist_download_pubmlst" version="0.1-SNAPSHOT">
+  <data_manager tool_file="data_manager_mentalist_download_pubmlst/data_manager/mentalist_download_pubmlst.xml" id="mentalist_download_pubmlst" version="0.1.8">
     <data_table name="mentalist_databases">
       <output>
         <column name="value" />
@@ -45,7 +45,7 @@
       </output>
     </data_table>
   </data_manager>
-  <data_manager tool_file="data_manager_mentalist_download_enterobase/data_manager/mentalist_download_enterobase.xml" id="mentalist_download_enterobase" version="0.1-SNAPSHOT">
+  <data_manager tool_file="data_manager_mentalist_download_enterobase/data_manager/mentalist_download_enterobase.xml" id="mentalist_download_enterobase" version="0.1.8">
     <data_table name="mentalist_databases">
       <output>
         <column name="value" />

galaxy/data_managers/data_manager_mentalist_build_db/data_manager/mentalist_build_db.xml

@@ -1,7 +1,7 @@
 <?xml version="1.0"?>
-<tool id="mentalist_build_db" name="MentaLiST Build DB" tool_type="manage_data" version="0.1-SNAPSHOT">
+<tool id="mentalist_build_db" name="MentaLiST Build DB" tool_type="manage_data" version="0.1.8">
   <requirements>
-    <requirement type="package" version="0.1_SNAPSHOT">mentalist</requirement>
+    <requirement type="package" version="0.1.8">mentalist</requirement>
   </requirements>
   <command detect_errors="exit_code">
   <![CDATA[

galaxy/data_managers/data_manager_mentalist_download_cgmlst/data_manager/mentalist_download_cgmlst.xml

@@ -1,7 +1,7 @@
 <?xml version="1.0"?>
-<tool id="mentalist_download_cgmlst" name="MentaLiST Download from cgMLST" tool_type="manage_data" version="0.1-SNAPSHOT">
+<tool id="mentalist_download_cgmlst" name="MentaLiST Download from cgMLST" tool_type="manage_data" version="0.1.8">
   <requirements>
-    <requirement type="package" version="0.1_SNAPSHOT">mentalist</requirement>
+    <requirement type="package" version="0.1.8">mentalist</requirement>
   </requirements>
   <command detect_errors="exit_code">
     <![CDATA[

galaxy/data_managers/data_manager_mentalist_download_enterobase/data_manager/mentalist_download_enterobase.xml

@@ -1,7 +1,7 @@
 <?xml version="1.0"?>
-<tool id="mentalist_download_enterobase" name="MentaLiST Download from Enterobase" tool_type="manage_data" version="0.1-SNAPSHOT">
+<tool id="mentalist_download_enterobase" name="MentaLiST Download from Enterobase" tool_type="manage_data" version="0.1.8">
   <requirements>
-    <requirement type="package" version="0.1_SNAPSHOT">mentalist</requirement>
+    <requirement type="package" version="0.1.8">mentalist</requirement>
   </requirements>
   <command detect_errors="exit_code">
     <![CDATA[

galaxy/data_managers/data_manager_mentalist_download_pubmlst/data_manager/mentalist_download_pubmlst.xml

@@ -1,7 +1,7 @@
 <?xml version="1.0"?>
-<tool id="mentalist_download_pubmlst" name="MentaLiST Download from pubMLST" tool_type="manage_data" version="0.1-SNAPSHOT">
+<tool id="mentalist_download_pubmlst" name="MentaLiST Download from pubMLST" tool_type="manage_data" version="0.1.8">
   <requirements>
-    <requirement type="package" version="0.1_SNAPSHOT">mentalist</requirement>
+    <requirement type="package" version="0.1.8">mentalist</requirement>
   </requirements>
   <command detect_errors="exit_code">
     <![CDATA[

galaxy/test-data/mentalist_tree/med_output_00.nwk

@@ -0,0 +1 @@
+(((((sample-55:41.71429,sample-46:41.28571)Inner36:2.08594,(sample-12:42.79847,sample-6:42.20153)Inner50:1.41406)Inner73:1.17639,(((((sample-89:39.37228,sample-72:40.62772)Inner7:2.88906,(sample-45:40.87658,sample-41:41.12342)Inner20:2.11094)Inner59:0.95241,(sample-27:41.71186,sample-24:41.28814)Inner40:2.29759)Inner77:0.51510,(sample-80:41.63672,(sample-82:40.88750,sample-21:41.11250)Inner19:0.86328)Inner35:2.70365)Inner84:0.48215,(((sample-53:39.58763,sample-9:39.41237)Inner2:3.35795,(sample-4:41.68852,sample-3:42.31148)Inner38:1.14205)Inner55:1.86063,(((sample-99:43.18750,(sample-69:40.72472,sample-11:40.27528)Inner10:2.31250)Inner57:0.78292,(sample-59:40.65000,sample-5:40.35000)Inner9:3.21708)Inner71:0.49007,((sample-64:42.29009,sample-51:41.70991)Inner46:1.50781,((sample-92:39.63235,sample-40:40.36765)Inner14:2.19477,(sample-36:41.94129,sample-0:42.05871)Inner33:1.30523)Inner56:0.99219)Inner72:0.29118)Inner75:0.68624)Inner85:0.26394)Inner90:0.19666)Inner95:0.08868,(((sample-98:41.20513,sample-10:40.79487)Inner21:2.48769,((sample-97:39.56720,sample-76:39.43280)Inner6:1.17105,sample-7:41.32895)Inner23:2.51231)Inner66:1.14558,((sample-74:41.75746,sample-60:41.24254)Inner32:2.66667,(sample-57:40.47159,sample-1:40.52841)Inner11:3.58333)Inner81:0.38567)Inner86:0.41473)Inner97:0.06049,((((((sample-85:40.63889,sample-79:40.36111)Inner18:2.67318,(sample-34:41.67500,(sample-61:41.34740,sample-29:40.65260)Inner22:0.82500)Inner39:1.57682)Inner63:0.86737,(sample-28:42.59722,sample-26:42.40278)Inner45:1.94513)Inner78:0.65710,((sample-68:41.90893,sample-25:41.09107)Inner29:2.04375,(sample-50:42.40701,(sample-38:39.78646,sample-20:39.21354)Inner3:3.09299)Inner58:0.45625)Inner64:1.28040)Inner91:0.18081,(((sample-75:43.23395,(sample-63:42.24892,(sample-49:40.47632,sample-47:39.52368)Inner4:2.25108)Inner41:1.01605)Inner62:0.40142,(sample-67:40.70640,sample-22:41.29360)Inner13:2.78608)Inner65:1.37617,(((sample-73:40.43750,sample-42:40.56250)Inner15:2.78987,(sample-54:40.93373,sample-33:41.06627)Inner16:2.71013)Inner70:0.76270,((sample-95:41.60769,sample-43:41.39231)Inner34:1.04255,(sample-39:39.71809,sample-17:40.28191)Inner5:2.70745)Inner52:1.79980)Inner83:0.34258)Inner89:0.30845)Inner94:0.14628,(((sample-87:41.83004,(sample-48:41.19366,sample-19:40.80634)Inner28:0.66996)Inner42:2.31867,((sample-91:43.22596,(sample-81:42.31971,sample-30:42.68029)Inner47:0.77404)Inner60:0.92154,(sample-90:40.65854,sample-8:40.34146)Inner17:2.82846)Inner76:0.33758)Inner80:0.63334,(((sample-52:42.86528,(sample-71:41.14062,sample-16:41.85938)Inner27:1.63472)Inner54:1.13031,(sample-88:42.35491,(sample-93:39.14796,sample-13:38.85204)Inner1:3.14509)Inner43:1.93219)Inner74:0.91706,(((sample-83:40.94000,sample-70:41.06000)Inner24:2.54713,(sample-58:42.45098,(sample-32:40.81926,sample-15:41.18074)Inner25:1.04902)Inner48:1.32787)Inner68:0.77832,((sample-84:40.79281,sample-56:41.20719)Inner26:1.93092,(sample-18:42.71250,sample-2:42.28750)Inner49:0.81908)Inner61:1.22168)Inner79:0.55950)Inner92:0.07174)Inner93:0.17696)Inner96:0.08099,(((sample-96:40.31319,sample-66:39.68681)Inner8:3.27246,(sample-62:42.03226,sample-35:40.96774)Inner37:1.72754)Inner67:1.19815,((sample-31:42.10677,(sample-44:40.49425,sample-23:40.50575)Inner12:1.89323)Inner51:1.99544,(((sample-78:42.29891,sample-77:42.70109)Inner53:0.82760,(sample-94:40.98713,sample-65:41.01287)Inner31:2.17240)Inner69:0.84881,(sample-86:41.76705,(sample-37:41.45833,sample-14:41.54167)Inner30:0.73295)Inner44:2.08869)Inner82:0.34831)Inner87:0.14560)Inner88:0.30133)Inner98:0.00000;

galaxy/test-data/mentalist_tree/small_input_00.tsv

@@ -0,0 +1,11 @@
+	sample-0	sample-1	sample-2	sample-3	sample-4	sample-5	sample-6	sample-7	sample-8	sample-9
+sample-0	0	10	10	9	10	10	8	9	10	10
+sample-1	10	0	9	9	8	7	9	7	9	9
+sample-2	10	9	0	10	9	10	9	10	10	10
+sample-3	9	9	10	0	10	10	8	10	9	8
+sample-4	10	8	9	10	0	7	9	10	10	6
+sample-5	10	7	10	10	7	0	10	8	9	10
+sample-6	8	9	9	8	9	10	0	10	8	7
+sample-7	9	7	10	10	10	8	10	0	9	9
+sample-8	10	9	10	9	10	9	8	9	0	8
+sample-9	10	9	10	8	6	10	7	9	8	0

galaxy/test-data/mentalist_tree/small_output_00.nwk

@@ -0,0 +1 @@
+((sample-8:4.42708,(sample-3:4.32812,(sample-6:3.42500,sample-0:4.57500)Inner4:0.17188)Inner5:0.32292)Inner6:0.17969,(sample-2:5.13281,(sample-7:3.87500,(sample-5:3.78571,sample-1:3.21429)Inner2:0.12500)Inner3:0.86719)Inner7:0.05469,(sample-9:2.87500,sample-4:3.12500)Inner1:1.32031)Inner8:0.00000;

galaxy/tools/mentalist_call/mentalist_call.xml

@@ -1,6 +1,6 @@
-<tool id="mentalist_call" name="MentaLiST MLST Analysis" version="0.1-SNAPSHOT">
+<tool id="mentalist_call" name="MentaLiST MLST Analysis" version="0.1.8">
   <requirements>
-    <requirement type="package" version="0.1_SNAPSHOT">mentalist</requirement>
+    <requirement type="package" version="0.1.8">mentalist</requirement>
   </requirements>
   <command detect_errors="exit_code"><![CDATA[
     mentalist

galaxy/tools/mentalist_distance/mentalist_distance.xml

@@ -1,4 +1,4 @@
-<tool id="mentalist_distance" name="MentaLiST Distance Matrix" version="0.1-SNAPSHOT">
+<tool id="mentalist_distance" name="MentaLiST Distance Matrix" version="0.1.8">
   <requirements>
     <requirement type="package" version="3.4.1">r-base</requirement>
   </requirements>
@@ -15,16 +15,16 @@
   </outputs>
   <tests>
     <test>
-      <param name="input" value="small_input_00.tsv" />
-      <output name="output" file="small_output_00.tsv" />
+      <param name="input" value="mentalist_distance/small_input_00.tsv" />
+      <output name="output" file="mentalist_distance/small_output_00.tsv" />
     </test>
     <test>
-      <param name="input" value="med_input_00.tsv" />
-      <output name="output" file="med_output_00.tsv" />
+      <param name="input" value="mentalist_distance/med_input_00.tsv" />
+      <output name="output" file="mentalist_distance/med_output_00.tsv" />
     </test>
     <test>
-      <param name="input" value="large_input_00.tsv" />
-      <output name="output" file="large_output_00.tsv" />
+      <param name="input" value="mentalist_distance/large_input_00.tsv" />
+      <output name="output" file="mentalist_distance/large_output_00.tsv" />
     </test>
   </tests>
   <help><![CDATA[

galaxy/tools/mentalist_tree/mentalist_tree

@@ -0,0 +1,66 @@
+#!/usr/bin/env python
+
+import sys
+import csv
+import numpy as np
+
+import Bio.Phylo
+from Bio.Phylo.TreeConstruction import DistanceMatrix, DistanceTreeConstructor
+
+def usage():
+  print("usage: mentalist_tree <input.tsv>\n")
+
+def process_input_matrix(input_matrix):
+    """ Converts an array-of-arrays containting sample IDs and distances
+        into a BioPython DistanceMatrix object
+    """
+    input_matrix.pop(0)
+    sample_names = [row[0] for row in input_matrix]
+    for row in input_matrix:
+        row.pop(0)
+    distance_matrix = []
+    for input_matrix_row in input_matrix:
+        distance_matrix.append([int(i) for i in input_matrix_row])
+    """ np.tril() converts a matrix like this: [[0 1 2]
+                                                [1 0 1]
+                                                [2 1 0]]
+        ...into this: [[0 0 0]
+                       [1 0 0]
+                       [2 1 0]]
+        ...but what we need to pass to DistanceMatrix() is this: [[0]
+                                                                  [1 0]
+                                                                  [2 1 0]]
+        ...so that's what the (somewhat cryptic) code below does.
+    """
+    distance_matrix = np.tril(np.array(distance_matrix))
+    num_rows = distance_matrix.shape[0]
+    """ masking the distance matrix with tril_indices gives a linearized
+        distance matrix [0 1 0 2 1 0] that we need to re-construct into [[0], [1, 0], [2, 1, 0]]
+    """
+    lower_triangular_idx_mask = np.tril_indices(num_rows)
+    linear_distance_matrix = distance_matrix[lower_triangular_idx_mask]
+    distance_matrix = []
+    min = 0
+    max = 1
+    for i in range(num_rows):
+        distance_matrix.append(linear_distance_matrix[min:max].tolist())
+        min = max
+        max = max + (i + 2) 
+    distance_matrix = DistanceMatrix(names=sample_names, matrix=distance_matrix)
+    return distance_matrix
+
+def main():
+    if len(sys.argv) < 2:
+        usage()
+        sys.exit(1)
+
+    input_file = sys.argv[1]
+    reader = csv.reader(open(input_file, "r"), delimiter="\t")
+    input_matrix = list(reader)
+    distance_matrix = process_input_matrix(input_matrix)
+    constructor = DistanceTreeConstructor()
+    tree = constructor.nj(distance_matrix)
+    Bio.Phylo.write(tree, sys.stdout, 'newick')
+
+if __name__ == '__main__':
+    main()

galaxy/tools/mentalist_tree/mentalist_tree.xml

@@ -0,0 +1,31 @@
+<tool id="mentalist_tree" name="MentaLiST Tree" version="0.1.8">
+  <requirements>
+    <requirement type="package" version="1.70">biopython</requirement>
+  </requirements>
+  <command detect_errors="exit_code"><![CDATA[
+    $__tool_directory__/mentalist_tree
+    '${input}'
+    > '${output}'
+  ]]></command>
+  <inputs>
+    <param name="input" type="data" format="tabular" label="MentaLiST Distance Matrix" help=""/>
+  </inputs>
+  <outputs>
+    <data name="output" format="txt"/>
+  </outputs>
+  <tests>
+    <test>
+      <param name="input" value="mentalist_tree/small_input_00.tsv" />
+      <output name="output" file="mentalist_tree/small_output_00.nwk" />
+    </test>
+    <test>
+      <param name="input" value="mentalist_tree/med_input_00.tsv" />
+      <output name="output" file="mentalitst_tree/med_output_00.nwk" />
+    </test>
+  </tests>
+  <help><![CDATA[
+    mentalist_distance <input.tsv>
+  ]]></help>
+  <citations>
+  </citations>
+</tool>