# HG changeset patch
# User rnateam
# Date 1481891729 18000
# Node ID 15bd4fb05e5c73dcd6e522602cece70d6046ae93

planemo upload for repository https://github.com/eteriSokhoyan/galaxytools/tree/branchForIterations/tools/GraphClust/LocARNAGraphClust commit 21aaee40723b5341b4236edeb0e72995c2054053

diff -r 000000000000 -r 15bd4fb05e5c locarna_best_subtree.xml
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/locarna_best_subtree.xml	Fri Dec 16 07:35:29 2016 -0500
@@ -0,0 +1,190 @@
+<tool id="locarna_best_subtree" name="locarna_best_subtree" version="0.1.0" >
+	<requirements>
+		<requirement type="package" version="0.1">graphclust-wrappers</requirement>
+		<requirement type="package" version='1.8.10'>locarna</requirement>
+		<requirement type="package" version='2.1'>rnaz</requirement>
+		<requirement type="package" version='0.07'>perl-math-round</requirement>
+	</requirements>
+	<stdio>
+		<exit_code range="1:" />
+	</stdio>
+	<command>
+		<![CDATA[
+
+		    'locARNAGraphClust.pl'  '$center_fa_file' '$tree_file' '$tree_matrix' $p $max_diff_am $tau $max_diff '' '$data_map' $plfold_minlen
+        ]]>
+	</command>
+	<inputs>
+		<param type="data" name="center_fa_file" label="centers" format="fa, fasta" help="fasta format" />
+		<param type="data" name="tree_file" label="trees" format="text" help="text format" />
+		<param type="data" name="tree_matrix" label="tree_matrix" format="text" help="text format" />
+		<param type="data" name="data_map" label="data_map" format="txt" help="text format" />
+		<param name="p" type="float" value="0.001" size="5" label="minimal probability" help="-p"/>
+		<param name="max_diff_am" type="integer" value="50" size="5" label=" maximal difference for sizes of matched arcs" help="--max-diff-am"/>
+		<param name="tau" type="integer" value="50" size="5" label="Tau factor in percent" help="--tau"/>
+		<param name="max_diff" type="integer" value="100" size="5" label="maximal difference for alignment traces" help="--max-diff"/>
+		<param name="plfold_minlen" type="integer" value="210" size="5" label="Minimal length of a sequences for which RNAplfold is used" />
+	</inputs>
+	<outputs>
+
+		<data name="model_tree_stk" format="stockholm" label="model.tree.stk" from_work_dir="MODEL/best_subtree.aln" />
+	</outputs>
+	<tests>
+		<test>
+			<param name="tree_file" value="1.1.tree"/>
+			<param name="center_fa_file" value="1.1.center.fa"/>
+			<param name="data_map" value="data.map"/>
+			<param name="tree_matrix" value="1.1.matrix.tree"/>
+			<param name="p" value="0.001"/>
+			<param name="max-diff-am" value="50"/>
+			<param name="tau" value="50"/>
+			<param name="max-diff-am" value="100"/>
+			<output name="model_tree_stk" file="best_subtree.aln"/>
+		</test>
+	</tests>
+	<help>
+		<![CDATA[
+**What it does**
+
+MLocARNA computes a multiple sequence-structure alignment of RNA sequences.
+It uses *treefile* - file with guide tree in NEWICK format. The given tree is used as guide tree for the progressive alignment.
+This saves the calculation of pairwise all-vs-all similarities and construction of the guide tree.
+
+
+
+]]>
+	</help>
+	<citations>
+		<citation type="bibtex">@inproceedings{costa2010fast,
+        title={Fast neighborhood subgraph pairwise distance kernel},
+        author={Costa, Fabrizio and De Grave, Kurt},
+        booktitle={Proceedings of the 26th International Conference on Machine Learning},
+        pages={255--262},
+        year={2010},
+        organization={Omnipress}
+      }
+      </citation>
+			<citation type="bibtex">@Article{Will_Joshi_Hofacker-LocAR_Accur_bound-2012,
+  author =	 {Will, Sebastian and Joshi, Tejal and Hofacker, Ivo L. and
+                  Stadler, Peter F. and Backofen, Rolf},
+  title =	 {{LocARNA}-{P}: {Accurate} boundary prediction and improved
+                  detection of structural {RNAs}},
+  journal =	 {RNA},
+  year =	 {2012},
+  volume =	 {18},
+  number =	 {5},
+  pages =	 {900-14},
+  user =	 {will},
+  pmid =	 {22450757},
+  doi = 	 {10.1261/rna.029041.111},
+  issn = 	 {1469-9001},
+  issn = 	 {1355-8382},
+  abstract =	 {Current genomic screens for noncoding RNAs (ncRNAs) predict
+                  a large number of genomic regions containing potential
+                  structural ncRNAs. The analysis of these data requires
+                  highly accurate prediction of ncRNA boundaries and
+                  discrimination of promising candidate ncRNAs from weak
+                  predictions. Existing methods struggle with these goals
+                  because they rely on sequence-based multiple sequence
+                  alignments, which regularly misalign RNA structure and
+                  therefore do not support identification of structural
+                  similarities. To overcome this limitation, we compute
+                  columnwise and global reliabilities of alignments based on
+                  sequence and structure similarity; we refer to these
+                  structure-based alignment reliabilities as STARs. The
+                  columnwise STARs of alignments, or STAR profiles, provide a
+                  versatile tool for the manual and automatic analysis of
+                  ncRNAs. In particular, we improve the boundary prediction of
+                  the widely used ncRNA gene finder RNAz by a factor of 3 from
+                  a median deviation of 47 to 13 nt. Post-processing RNAz
+                  predictions, LocARNA-P's STAR score allows much stronger
+                  discrimination between true- and false-positive predictions
+                  than RNAz's own evaluation. The improved accuracy, in this
+                  scenario increased from AUC 0.71 to AUC 0.87, significantly
+                  reduces the cost of successive analysis steps. The
+                  ready-to-use software tool LocARNA-P produces
+                  structure-based multiple RNA alignments with associated
+                  columnwise STARs and predicts ncRNA boundaries. We provide
+                  additional results, a web server for LocARNA/LocARNA-P, and
+                  the software package, including documentation and a pipeline
+                  for refining screens for structural ncRNA, at
+                  http://www.bioinf.uni-freiburg.de/Supplements/LocARNA-P/.}
+}
+				</citation>
+				<citation type="bibtex">@Article{Will:etal:_infer_non_codin_rna_famil:PLOS2007,
+  author =	 {Sebastian Will and Kristin Reiche and Ivo L. Hofacker and
+                  Peter F. Stadler and Rolf Backofen},
+  title =	 {Inferring Non-Coding {RNA} Families and Classes by Means of
+                  Genome-Scale Structure-Based Clustering},
+  journal =	 {PLoS Comput Biol},
+  year =	 2007,
+  volume =       {3},
+  number =       {4},
+  pages =        {e65},
+  issn =         {1553-7358},
+  issn =         {1553-734X},
+  pmid =         {17432929},
+  doi =          {10.1371/journal.pcbi.0030065},
+  user =	 {will},
+  abstract =	 {The RFAM database defines families of ncRNAs by means of
+                  sequence similarities that are sufficientto establish
+                  homology. In some cases, such as microRNAs, box H/ACA
+                  snoRNAs, functional commonalities define classes of RNAs
+                  that are characterized by structural similarities, and
+                  typically consist ofmultiple RNA families. Recent advances
+                  in high-throughput transcriptomics and comparative genomics
+                  have produced very large sets of putative non-coding RNAs
+                  and regulatory RNA signals. For many ofthem, evidence for
+                  stabilizing selection acting on their secondary structures
+                  has been derived, and at least approximate models of their
+                  structures have been computed. The overwhelming majority of
+                  these hypo-thetical RNAs cannot be assigned to established
+                  families or classes. We present here a structure-based
+                  clustering approach that is capable of extracting putative
+                  RNA classesfrom genome-wide surveys for structured RNAs. The
+                  LocARNA tool implements a novel variant of theSankoff
+                  algorithm that is sufficiently fast to deal with several
+                  thousand candidate sequences. The method is also robust
+                  against false positive predictions, i.e., a contamination of
+                  the input data with unstructured ornon-conserved
+                  sequences. We have successfully tested the LocARNA-based
+                  clustering approach on the sequences of the
+                  RFAM-seedalignments. Furthermore, we have applied it to a
+                  previously published set of 3332 predicted structured
+                  elements in the Ciona intestinalis genomes (Missal et al.,
+                  Bioinformatics 21(S2), i77-i78). In addition torecovering
+                  e.g. tRNAs as a structure-based class, the method identifies
+                  several RNA families, including microRNA and snoRNA
+                  candidates, and suggests several novel classes of ncRNAs for
+                  which to-date norepresentative has been experimentally
+                  characterized.}
+}
+
+					</citation>
+					<citation type="bibtex">@Article{Smith:Heyne:Richter:Freib_RNA_Tools:NAR2010,
+  author =	 {Smith, Cameron and Heyne, Steffen and Richter, Andreas S.
+                  and Will, Sebastian and Backofen, Rolf},
+  title =	 {Freiburg {RNA} {Tools}: a web server integrating {IntaRNA},
+                  {ExpaRNA} and {LocARNA}},
+  journal =	 NAR,
+  year =	 {2010},
+  volume =	 {38 Suppl},
+  number =	 {},
+  pages =	 {W373-7},
+  user =	 {arichter},
+  pmid =	 {20444875},
+  doi = 	 {10.1093/nar/gkq316},
+  issn = 	 {0305-1048},
+  issn =	 {1362-4962},
+  abstract =	 {The Freiburg RNA tools web server integrates three tools
+                  for the advanced analysis of RNA in a common web-based user
+                  interface. The tools IntaRNA, ExpaRNA and LocARNA support
+                  the prediction of RNA-RNA interaction, exact RNA matching
+                  and alignment of RNA, respectively. The Freiburg RNA tools
+                  web server and the software packages of the stand-alone
+                  tools are freely accessible at
+                  http://rna.informatik.uni-freiburg.de.}
+}
+						</citation>
+	</citations>
+</tool>
diff -r 000000000000 -r 15bd4fb05e5c test-data/1.1.center.fa
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/test-data/1.1.center.fa	Fri Dec 16 07:35:29 2016 -0500
@@ -0,0 +1,40 @@
+>7 SEQ7#1#83#+ ORIGID RF00005_rep.12_AC108081.2/59868-59786_7 ORIGHEAD RF00005_rep.12
+GUCAGGAUGGCCGAGCGGUCUAAGGCGCUGCGUUCAGGUCGCAGUCUCCCCUGGAGGCGUGGGUUCGAAUCCCACUUCUGACA
+>9 SEQ9#1#73#+ ORIGID RF00005_rep.14_AL021808.2/65570-65498_9 ORIGHEAD RF00005_rep.14
+GCUUCUGUAGUGUAGUGGUUAUCACGUUCGCCUCACACGCGAAAGGUCCCCGGUUCGAAACCGGGCAGAAGCA
+>10 SEQ10#1#73#+ ORIGID RF00005_rep.15_AC008443.10/42590-42518_10 ORIGHEAD RF00005_rep.15
+GCCCGGCUAGCUCAGUCGGUAGAGCAUGAGACUCUUAAUCUCAGGGUCGUGGGUUCGAGCCCCACGUUGGGCG
+>15 SEQ15#1#73#+ ORIGID RF00005_rep.1_AC005329.1/7043-6971_15 ORIGHEAD RF00005_rep.1
+GCCGAAAUAGCUCAGUUGGGAGAGCGUUAGACUGAAGAUCUAAAGGUCCCUGGUUCGAUCCCGGGUUUCGGCA
+>16 SEQ16#1#72#+ ORIGID RF00005_rep.20_AL671879.2/100356-100285_16 ORIGHEAD RF00005_rep.20
+GGGGAUGUAGCUCAGUGGUAGAGCGCAUGCUUCGCAUGUAUGAGGCCCCGGGUUCGAUCCCCGGCAUCUCCA
+>17 SEQ17#1#71#+ ORIGID RF00005_rep.21_AL355149.13/15278-15208_17 ORIGHEAD RF00005_rep.21
+GCAUUGGUGGUUCAGUGGUAGAAUUCUCGCCUCCCACGCGGGAGACCCGGGUUCAAUUCCCGGCCAAUGCA
+>18 SEQ18#1#72#+ ORIGID RF00005_rep.22_AL590385.23/26487-26416_18 ORIGHEAD RF00005_rep.22
+GCGUUGGUGGUAUAGUGGUGAGCAUAGCUGCCUUCCAAGCAGUUGACCCGGGUUCGAUUCCCGGCCAACGCA
+>23 SEQ23#1#74#+ ORIGID RF00005_rep.27_AL352978.6/119697-119770_23 ORIGHEAD RF00005_rep.27
+GGCCGGUUAGCUCAGUUGGUUAGAGCGUGGUGCUAAUAACGCCAAGGUCGCGGGUUCGAUCCCCGUACGGGCCA
+>28 SEQ28#1#71#+ ORIGID RF00005_rep.31_AC092686.3/29631-29561_28 ORIGHEAD RF00005_rep.31
+GCAUUGGUGGUUCAGUGGUAGAAUUCUCGCCUGCCACGCGGGAGGCCCGGGUUCGAUUCCCGGCCAAUGCA
+>30 SEQ30#1#72#+ ORIGID RF00005_rep.33_AC018638.5/4694-4623_30 ORIGHEAD RF00005_rep.33
+GGCUCGUUGGUCUAGGGGUAUGAUUCUCGCUUAGGGUGCGAGAGGUCCCGGGUUCAAAUCCCGGACGAGCCC
+>31 SEQ31#1#73#+ ORIGID RF00005_rep.34_AC008443.10/43006-42934_31 ORIGHEAD RF00005_rep.34
+GUUUCCGUAGUGUAGUGGUUAUCACGUUCGCCUCACACGCGAAAGGUCCCCGGUUCGAAACCGGGCGGAAACA
+>32 SEQ32#1#73#+ ORIGID RF00005_rep.35_AC005783.1/27398-27326_32 ORIGHEAD RF00005_rep.35
+GUUUCCGUAGUGUAGCGGUUAUCACAUUCGCCUCACACGCGAAAGGUCCCCGGUUCGAUCCCGGGCGGAAACA
+>33 SEQ33#1#72#+ ORIGID RF00005_rep.36_AC007298.17/145366-145295_33 ORIGHEAD RF00005_rep.36
+UCCUCGUUAGUAUAGUGGUGAGUAUCCCCGCCUGUCACGCGGGAGACCGGGGUUCGAUUCCCCGACGGGGAG
+>35 SEQ35#1#72#+ ORIGID RF00005_rep.38_J00309.1/356-427_35 ORIGHEAD RF00005_rep.38
+UCCCUGGUGGUCUAGUGGCUAGGAUUCGGCGCUUUCACCGCCGCGCCCCGGGUUCGAUUCCCGGCCAGGAAU
+>37 SEQ37#1#82#+ ORIGID RF00005_rep.3_Z54587.1/126-45_37 ORIGHEAD RF00005_rep.3
+GGUAGCGUGGCCGAGCGGUCUAAGGCGCUGGAUUUAGGCUCCAGUCUCUUCGGAGGCGUGGGUUCGAAUCCCACCGCUGCCA
+>46 SEQ46#1#72#+ ORIGID RF00005_rep.5_AL590385.23/26129-26058_46 ORIGHEAD RF00005_rep.5
+UCCCUGGUGGUCUAGUGGUUAGGAUUCGGCGCUCUCACCGCCGCGGCCCGGGUUCGAUUCCCGGUCAGGGAA
+>51 SEQ51#1#88#+ ORIGID RF00006_rep.0_AF045145.1/1-88_51 ORIGHEAD RF00006_rep.0
+GGCUGGCUUUAGCUCAGCGGUUACUUCGCGUGUCAUCAAACCACCUCUCUGGGUUGUUCGAGACCCGCGGGCGCUCUCCAGCCCUCUU
+>52 SEQ52#1#101#+ ORIGID RF00006_rep.1_AC005219.1/49914-50014_52 ORIGHEAD RF00006_rep.1
+GGGUCGGAGUUAGCUCAAGCGGUUACCUCCUCAUGCCGGACUUUCUAUCUGUCCAUCUCUGUGCUGGGGUUCGAGACCCGCGGGUGCUUACUGACCCUUUU
+>53 SEQ53#1#98#+ ORIGID RF00006_rep.2_AF045143.1/1-98_53 ORIGHEAD RF00006_rep.2
+GGCUGGCUUUAGCUCAGCGGUUACUUCGACAGUUCUUUAAUUGAAACAAGCAACCUGUCUGGGUUGUUCGAGACCCGCGGGCGCUCUCCAGUCCUUUU
+>54 SEQ54#1#88#+ ORIGID RF00006_rep.3_AF045144.1/1-88_54 ORIGHEAD RF00006_rep.3
+GGCUGGCUUUAGCUCAGCGGUUACUUCGAGUACAUUGUAACCACCUCUCUGGGUGGUUCGAGACCCGCGGGUGCUUUCCAGCUCUUUU
diff -r 000000000000 -r 15bd4fb05e5c test-data/1.1.matrix.tree
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/test-data/1.1.matrix.tree	Fri Dec 16 07:35:29 2016 -0500
@@ -0,0 +1,20 @@
+0.000000  0.399566  0.449067  0.416346  0.422581  0.421691  0.420328  0.433314  0.421864  0.415656  0.388584  0.396080  0.395237  0.425138  0.525331  0.428431  0.423873  0.426227  0.430509  0.421103  
+0.399566  0.000000  0.413478  0.410598  0.415992  0.448487  0.444776  0.413799  0.454424  0.422079  0.758362  0.628040  0.449935  0.432423  0.395294  0.432108  0.413869  0.394038  0.410417  0.404625  
+0.449067  0.413478  0.000000  0.447753  0.452406  0.432149  0.429802  0.446808  0.434943  0.437722  0.401281  0.400204  0.406763  0.418486  0.433646  0.417759  0.427363  0.424735  0.425622  0.422313  
+0.416346  0.410598  0.447753  0.000000  0.432407  0.413443  0.406226  0.453570  0.424393  0.406074  0.403132  0.413663  0.412748  0.409772  0.411737  0.422837  0.408136  0.419974  0.412883  0.409892  
+0.422581  0.415992  0.452406  0.432407  0.000000  0.445562  0.440880  0.457374  0.453192  0.448142  0.396150  0.415425  0.416902  0.440906  0.424893  0.433323  0.417320  0.420319  0.412897  0.417445  
+0.421691  0.448487  0.432149  0.413443  0.445562  0.000000  0.473887  0.422858  0.692833  0.442948  0.434741  0.428054  0.474703  0.466744  0.423109  0.468801  0.428776  0.430573  0.428577  0.424501  
+0.420328  0.444776  0.429802  0.406226  0.440880  0.473887  0.000000  0.423941  0.488936  0.425014  0.412393  0.417502  0.429722  0.471749  0.413970  0.438547  0.402300  0.405816  0.416556  0.412890  
+0.433314  0.413799  0.446808  0.453570  0.457374  0.422858  0.423941  0.000000  0.427433  0.436563  0.407055  0.414622  0.415155  0.429164  0.444031  0.426024  0.435089  0.420686  0.422545  0.422448  
+0.421864  0.454424  0.434943  0.424393  0.453192  0.692833  0.488936  0.427433  0.000000  0.434849  0.439028  0.434897  0.467830  0.489166  0.434846  0.473733  0.418076  0.412067  0.415370  0.421449  
+0.415656  0.422079  0.437722  0.406074  0.448142  0.442948  0.425014  0.436563  0.434849  0.000000  0.414830  0.419767  0.423499  0.443441  0.420867  0.440759  0.423357  0.422084  0.413487  0.413927  
+0.388584  0.758362  0.401281  0.403132  0.396150  0.434741  0.412393  0.407055  0.439028  0.414830  0.000000  0.826283  0.458338  0.420348  0.389570  0.431953  0.402238  0.389074  0.397992  0.391832  
+0.396080  0.628040  0.400204  0.413663  0.415425  0.428054  0.417502  0.414622  0.434897  0.419767  0.826283  0.000000  0.444033  0.419958  0.392449  0.423827  0.410142  0.395175  0.397606  0.395594  
+0.395237  0.449935  0.406763  0.412748  0.416902  0.474703  0.429722  0.415155  0.467830  0.423499  0.458338  0.444033  0.000000  0.431310  0.401528  0.447342  0.406519  0.423555  0.410966  0.407189  
+0.425138  0.432423  0.418486  0.409772  0.440906  0.466744  0.471749  0.429164  0.489166  0.443441  0.420348  0.419958  0.431310  0.000000  0.442879  0.626901  0.429861  0.419845  0.414696  0.426070  
+0.525331  0.395294  0.433646  0.411737  0.424893  0.423109  0.413970  0.444031  0.434846  0.420867  0.389570  0.392449  0.401528  0.442879  0.000000  0.452328  0.441888  0.432467  0.434868  0.431986  
+0.428431  0.432108  0.417759  0.422837  0.433323  0.468801  0.438547  0.426024  0.473733  0.440759  0.431953  0.423827  0.447342  0.626901  0.452328  0.000000  0.423858  0.419329  0.419532  0.421059  
+0.423873  0.413869  0.427363  0.408136  0.417320  0.428776  0.402300  0.435089  0.418076  0.423357  0.402238  0.410142  0.406519  0.429861  0.441888  0.423858  0.000000  0.474825  0.594380  0.571671  
+0.426227  0.394038  0.424735  0.419974  0.420319  0.430573  0.405816  0.420686  0.412067  0.422084  0.389074  0.395175  0.423555  0.419845  0.432467  0.419329  0.474825  0.000000  0.466968  0.489649  
+0.430509  0.410417  0.425622  0.412883  0.412897  0.428577  0.416556  0.422545  0.415370  0.413487  0.397992  0.397606  0.410966  0.414696  0.434868  0.419532  0.594380  0.466968  0.000000  0.548030  
+0.421103  0.404625  0.422313  0.409892  0.417445  0.424501  0.412890  0.422448  0.421449  0.413927  0.391832  0.395594  0.407189  0.426070  0.431986  0.421059  0.571671  0.489649  0.548030  0.000000  
diff -r 000000000000 -r 15bd4fb05e5c test-data/1.1.tree
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/test-data/1.1.tree	Fri Dec 16 07:35:29 2016 -0500
@@ -0,0 +1,1 @@
+(((((54:0.133216,(53:0.115952,51:0.115952):0.0172647):0.0413516,52:0.174568):0.0233911,(37:0.150476,7:0.150476):0.047483):0.00361419,(((23:0.184455,16:0.184455):0.0038835,10:0.188338):0.00251517,15:0.190853):0.01072):0.00340405,((((46:0.099691,35:0.099691):0.0793888,((28:0.066725,17:0.066725):0.105711,18:0.172436):0.00664407):0.0153606,30:0.19444):0.00190942,(33:0.187757,((32:0,31:0):0.066541,9:0.066541):0.121216):0.00859264):0.00862737):0
diff -r 000000000000 -r 15bd4fb05e5c test-data/best_subtree.aln
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/test-data/best_subtree.aln	Fri Dec 16 07:35:29 2016 -0500
@@ -0,0 +1,8 @@
+CLUSTAL W --- LocARNA 1.8.10
+
+18                 GCGUUGGUGGUAUAGUGGUGAGCAUAGCUGCCUUCCAAGCAGUUGA-CCCGGGUUCGAUUCCCGGCCAACGCA
+17                 GCAUUGGUGGUUCAGUGGU-AGAAUUCUCGCCUCCCACGCGGGAGA-CCCGGGUUCAAUUCCCGGCCAAUGCA
+28                 GCAUUGGUGGUUCAGUGGU-AGAAUUCUCGCCUGCCACGCGGGAGG-CCCGGGUUCGAUUCCCGGCCAAUGCA
+46                 UCCCUGGUGGUCUAGUGGUUAGGAUUCGGCGCUCUCACCGCCGCGG-CCCGGGUUCGAUUCCCGGUCAGGGAA
+35                 UCCCUGGUGGUCUAGUGGCUAGGAUUCGGCGCUUUCACCGCCGCGC-CCCGGGUUCGAUUCCCGGCCAGGAAU
+30                 GGCUCGUUGGUCUAGGGGU-AUGAUUCUCGCUUAGGGUGCGAGAGGUCCCGGGUUCAAAUCCCGGACGAGCCC
diff -r 000000000000 -r 15bd4fb05e5c test-data/data.map
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/test-data/data.map	Fri Dec 16 07:35:29 2016 -0500
@@ -0,0 +1,92 @@
+1 SEQ1#1#120#+
+2 SEQ2#1#118#+
+3 SEQ3#1#104#+
+4 SEQ4#1#73#+
+5 SEQ5#1#72#+
+6 SEQ6#1#66#+
+7 SEQ7#1#83#+
+8 SEQ8#1#70#+
+9 SEQ9#1#73#+
+10 SEQ10#1#73#+
+11 SEQ11#1#82#+
+12 SEQ12#1#82#+
+13 SEQ13#1#82#+
+14 SEQ14#1#73#+
+15 SEQ15#1#73#+
+16 SEQ16#1#72#+
+17 SEQ17#1#71#+
+18 SEQ18#1#72#+
+19 SEQ19#1#82#+
+20 SEQ20#1#72#+
+21 SEQ21#1#74#+
+22 SEQ22#1#72#+
+23 SEQ23#1#74#+
+24 SEQ24#1#73#+
+25 SEQ25#1#69#+
+26 SEQ26#1#72#+
+27 SEQ27#1#69#+
+28 SEQ28#1#71#+
+29 SEQ29#1#66#+
+30 SEQ30#1#72#+
+31 SEQ31#1#73#+
+32 SEQ32#1#73#+
+33 SEQ33#1#72#+
+34 SEQ34#1#68#+
+35 SEQ35#1#72#+
+36 SEQ36#1#73#+
+37 SEQ37#1#82#+
+38 SEQ38#1#65#+
+39 SEQ39#1#69#+
+40 SEQ40#1#66#+
+41 SEQ41#1#68#+
+42 SEQ42#1#69#+
+43 SEQ43#1#71#+
+44 SEQ44#1#69#+
+45 SEQ45#1#72#+
+46 SEQ46#1#72#+
+47 SEQ47#1#68#+
+48 SEQ48#1#71#+
+49 SEQ49#1#68#+
+50 SEQ50#1#73#+
+51 SEQ51#1#88#+
+52 SEQ52#1#101#+
+53 SEQ53#1#98#+
+54 SEQ54#1#88#+
+55 SEQ55#1#113#+
+56 SEQ56#1#95#+
+57 SEQ57#1#102#+
+58 SEQ58#1#84#+
+59 SEQ59#1#79#+
+60 SEQ60#1#87#+
+61 SEQ61#1#76#+
+62 SEQ62#1#62#+
+63 SEQ63#1#68#+
+64 SEQ64#1#62#+
+65 SEQ65#1#57#+
+66 SEQ66#1#71#+
+67 SEQ67#1#66#+
+68 SEQ68#1#64#+
+69 SEQ69#1#82#+
+70 SEQ70#1#68#+
+71 SEQ71#1#76#+
+72 SEQ72#1#126#+
+73 SEQ73#1#126#+
+74 SEQ74#1#127#+
+75 SEQ75#1#126#+
+76 SEQ76#1#79#+
+77 SEQ77#1#77#+
+78 SEQ78#1#80#+
+79 SEQ79#1#79#+
+80 SEQ80#1#79#+
+81 SEQ81#1#77#+
+82 SEQ82#1#81#+
+83 SEQ83#1#79#+
+84 SEQ84#1#79#+
+85 SEQ85#1#79#+
+86 SEQ86#1#79#+
+87 SEQ87#1#78#+
+88 SEQ88#1#81#+
+89 SEQ89#1#80#+
+90 SEQ90#1#66#+
+91 SEQ91#1#69#+
+92 SEQ92#1#69#+