.travis.yml

 language: python
-sudo: true
 python:
+  - "3.6"
+  - "3.5"
   - "2.7"
-# command to install dependencies
-virtualenv:
-  system_site_packages: true
-before_install:
-  - sudo apt-get install python-numpy python-scipy
 install:
-  - pip install biopython
-  - python setup.py build
+  - sudo apt-get update
+  - wget https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh -O miniconda.sh;
+  - bash miniconda.sh -b -p $HOME/miniconda
+  - export PATH="$HOME/miniconda/bin:$PATH"
+  - hash -r
+  - conda config --set always_yes yes --set changeps1 no
+  - conda update -q conda
+  # Useful for debugging any issues with conda
+  - conda info -a
+
+  - conda create -q -n test-environment python=$TRAVIS_PYTHON_VERSION numpy scipy biopython nose
+  - source activate test-environment
   - python setup.py install
-# command to run tests
 script:
   - bash .travis_test.sh

.travis_test.sh

-nosetests test/test_treetime.py
+cd test
+bash command_line_tests.sh
+OUT=$?
+if [ "$OUT" != 0 ]; then
+	exit 1
+fi
+nosetests test_treetime.py

README.md

@@ -6,7 +6,6 @@
 
 TreeTime provides routines for ancestral sequence reconstruction and the inference of molecular-clock phylogenies, i.e., a tree where all branches are scaled such that the locations of terminal nodes correspond to their sampling times and internal nodes are placed at the most likely time of divergence.
 
-TreeTime is implemented in Python 2.7 -- an experimental port to [Python 3 is available in a separate branch](https://github.com/neherlab/treetime/tree/py3).
 TreeTime aims at striking a compromise between sophisticated probabilistic models of evolution and fast heuristics. It implements GTR models of ancestral inference and branch length optimization, but takes the tree topology as given.
 To optimize the likelihood of time-scaled phylogenies, TreeTime uses an iterative approach that first infers ancestral sequences given the branch length of the tree, then optimizes the positions of unconstrained nodes on the time axis, and then repeats this cycle.
 The only topology optimization are (optional) resolution of polytomies in a way that is most (approximately) consistent with the sampling time constraints on the tree.
@@ -34,22 +33,22 @@ The package is designed to be used as a stand-alone tool or as a library used in
 
 
 ### Installation and prerequisites
-TreeTime is written in Python 2.7 and currently doesn't support Python 3.
-* The package depends on several python libraries:
-    - numpy, scipy: for all kind of mathematical operations as matrix operations, numerical integration, interpolation, minimization, etc.
 
-    - BioPython: for parsing multiple sequence alignments and all phylogenetic functionality
+TreeTime is compatible with Python 2.7 upwards and is tested on 2.7, 3.5, and 3.6.  It depends on several Python libraries:
 
-    - matplotlib: optional dependency for plotting
-  If you do not have these libraries, you can install them by typing in the terminal:
-    ```bash
-    $pip install numpy scipy biopython matplotlib
-    ```
+* numpy, scipy, pandas: for all kind of mathematical operations as matrix
+  operations, numerical integration, interpolation, minimization, etc.
 
-* To install the package, run `setup.py` script from the terminal:
-    ```bash
-    $python setup.py install
-    ```
+* BioPython: for parsing multiple sequence alignments and all phylogenetic
+  functionality
+
+* matplotlib: optional dependency for plotting
+
+You may install TreeTime and its dependencies by running
+
+    pip install .
+
+within this repository.
 
 You might need root privileges for system wide installation. Alternatively, you can simply use it TreeTime locally without installation. In this case, just download and unpack it, and then add the TreeTime folder to your $PYTHONPATH.
 

bin/ancestral_reconstruction.py

@@ -4,6 +4,7 @@ import numpy as np
 from treetime import TreeAnc, GTR
 from Bio import Phylo, AlignIO
 from Bio import __version__ as bioversion
+import sys
 
 if __name__=="__main__":
     ###########################################################################
@@ -85,7 +86,8 @@ if __name__=="__main__":
     ###########################################################################
     ### ANCESTRAL RECONSTRUCTION
     ###########################################################################
-    treeanc = TreeAnc(params.tree, aln=params.aln, gtr=gtr, verbose=4, fill_overhangs=not params.keep_overhangs)
+    treeanc = TreeAnc(params.tree, aln=params.aln, gtr=gtr, verbose=4,
+                      fill_overhangs=not params.keep_overhangs)
     treeanc.infer_ancestral_sequences('ml', infer_gtr=infer_gtr,
                                        marginal=params.marginal)
 
@@ -121,3 +123,5 @@ if __name__=="__main__":
     outtree_name = '.'.join(params.tree.split('/')[-1].split('.')[:-1])+'_mutation.nexus'
     Phylo.write(treeanc.tree, outtree_name, 'nexus')
     print("--- tree saved in nexus format as  \n\t %s\n"%outtree_name)
+
+    sys.exit(0)

bin/homoplasy_scanner.py

@@ -4,7 +4,7 @@ import numpy as np
 from treetime import TreeAnc, GTR
 from Bio import Phylo, AlignIO
 from Bio import __version__ as bioversion
-import os,shutil
+import os,shutil, sys
 
 if __name__=="__main__":
     ###########################################################################
@@ -29,7 +29,6 @@ if __name__=="__main__":
         "Example: '--gtr K80 --gtr_params kappa=0.2 pis=0.25,0.25,0.25,0.25'. See the exact definitions of "
         " the parameters in the GTR creation methods in treetime/nuc_models.py. Only nucleotide models supported at present")
 
-    parser.add_argument('--prot', default = False, action="store_true", help ="protein alignment")
     parser.add_argument('--zero_based', default = False, action='store_true', help='zero based SNP indexing')
     parser.add_argument('-n', default = 10, type=int, help='number of mutations/nodes that are printed to screen')
     parser.add_argument('--verbose', default = 1, type=int, help='verbosity of output 0-6')
@@ -39,6 +38,7 @@ if __name__=="__main__":
     ###########################################################################
     ### CHECK FOR TREE, build if not in place
     ###########################################################################
+
     if params.tree is None:
         from treetime.utils import tree_inference
         params.tree = os.path.basename(params.aln)+'.nwk'
@@ -47,6 +47,9 @@ if __name__=="__main__":
         tree_inference(params.aln, params.tree, tmp_dir = tmp_dir)
         if os.path.isdir(tmp_dir):
             shutil.rmtree(tmp_dir)
+    elif not os.path.isfile(params.tree):
+        print("Input tree file does not exist:", params.tree)
+        exit(1)
 
     ###########################################################################
     ### GTR SET-UP
@@ -216,3 +219,6 @@ if __name__=="__main__":
         for name, val in mutation_by_strain_sorted[:params.n]:
             if len(val):
                 print("\t%s\t%d"%(name, len(val)))
+
+
+    sys.exit(0)

bin/mugration.py

@@ -8,7 +8,7 @@ from Bio.Seq import Seq
 from Bio.Align import MultipleSeqAlignment
 from Bio import Phylo, AlignIO
 from Bio import __version__ as bioversion
-import os
+import os,sys
 
 if __name__=="__main__":
     ###########################################################################
@@ -23,7 +23,7 @@ if __name__=="__main__":
     parser.add_argument('--states', required = True, type=str, help ="csv or tsv file with discrete characters."
                                     "\n#name,country,continent\ntaxon1,micronesia,oceania\n...")
     parser.add_argument('--weights', type=str, help="csv or tsv file with probabilities of that a randomly sampled "
-                        "sequence has a particular state. E.g. population of different continents or countries. E.g.:"
+                        "sequence at equilibrium has a particular state. E.g. population of different continents or countries. E.g.:"
                         "\n#country,weight\nmicronesia,0.1\n...")
     # parser.add_argument('--migration', type=str, help="csv or tsv file with symmetric migration/transition rates "
     #                     "between states. For example passenger flow.")
@@ -79,7 +79,7 @@ if __name__=="__main__":
         tmp_weights = pd.read_csv(params.weights, sep='\t' if params.states[-3:]=='tsv' else ',',
                              skipinitialspace=True)
         weights = {row[0]:row[1] for ri,row in tmp_weights.iterrows()}
-        mean_weight = np.mean(weights.values())
+        mean_weight = np.mean(list(weights.values()))
         weights = np.array([weights[c] if c in weights else mean_weight for c in unique_states], dtype=float)
         weights/=weights.sum()
     else:
@@ -144,3 +144,5 @@ if __name__=="__main__":
     outtree_name = bname+'.mugration.nexus'
     Phylo.write(treeanc.tree, outtree_name, 'nexus')
     print("Saved annotated tree as:",outtree_name)
+
+    sys.exit(0)

bin/temporal_signal.py

@@ -4,6 +4,7 @@ import numpy as np
 from treetime import TreeTime
 from treetime.utils import DateConversion
 from Bio import Phylo, AlignIO
+import sys
 
 if __name__=="__main__":
     ###########################################################################
@@ -75,6 +76,9 @@ if __name__=="__main__":
           '\nthe substitution rate. The rate needs to be positive!'
           '\nNegative rates suggest an inappropriate root.\n\n')
 
+    print('\nThe estimated rate and tree correspond to a root date:\n')
+    print('\n--root-date:\t %3.2f\n\n'%(-d2d.intercept/d2d.clock_rate))
+
     if not params.keep_root:
         # write rerooted tree to file
         outtree_name = base_name+'_rerooted.newick'
@@ -108,6 +112,7 @@ if __name__=="__main__":
         plt.savefig(base_name+'_root_to_tip_regression.pdf')
         print("--- root-to-tip plot saved to  \n\t %s_root_to_tip_regression.pdf\n"%base_name)
 
+    sys.exit(0)
 
 
 

bin/timetree_inference.py

@@ -4,6 +4,7 @@ import numpy as np
 from treetime import TreeTime, GTR
 from Bio import Phylo, AlignIO
 from Bio import __version__ as bioversion
+import sys
 
 if __name__=="__main__":
     ###########################################################################
@@ -135,7 +136,7 @@ if __name__=="__main__":
     myTree.run(root=params.reroot, relaxed_clock=params.relax,
                resolve_polytomies=(not params.keep_polytomies),
                Tc=Tc, max_iter=params.max_iter,
-               use_input_branch_length = (not params.optimize_branch_length))
+               branch_lengths = 'joint' if params.optimize_branch_length else 'input')
 
     ###########################################################################
     ### OUTPUT and saving of results
@@ -193,3 +194,5 @@ if __name__=="__main__":
     Phylo.write(myTree.tree, outtree_name, 'nexus')
 
     print("--- tree saved in nexus format as  \n\t %s\n"%outtree_name)
+
+    sys.exit(0)

debian/changelog

+python-treetime (0.4.0-1) UNRELEASED; urgency=medium
+
+  * New upstream version
+  * Drop ancient X-Python-Version field
+  * Provide Python3 package, move tools also to Python3 package
+
+ -- Andreas Tille <tille@debian.org>  Fri, 29 Jun 2018 17:06:03 +0200
+
 python-treetime (0.2.4-1) unstable; urgency=medium
 
   * d/watch: Point to github since upstream has started using release tags

debian/control

@@ -8,11 +8,12 @@ Build-Depends: debhelper (>= 11~),
                dh-python,
                python-all,
                python-setuptools
+               python3-all,
+               python3-setuptools
 Standards-Version: 4.1.4
 Vcs-Browser: https://salsa.debian.org/med-team/python-treetime
 Vcs-Git: https://salsa.debian.org/med-team/python-treetime.git
 Homepage: https://github.com/neherlab/treetime
-X-Python-Version: >= 2.6
 
 Package: python-treetime
 Architecture: all
@@ -23,7 +24,7 @@ Depends: ${python:Depends},
          python-scipy,
          python-biopython,
          python-pandas
-Description: inference of time stamped phylogenies and ancestral reconstruction
+Description: inference of time stamped phylogenies and ancestral reconstruction (Python 2)
  TreeTime provides routines for ancestral sequence reconstruction and the
  maximum likelihoo inference of molecular-clock phylogenies, i.e., a tree
  where all branches are scaled such that the locations of terminal nodes
@@ -51,6 +52,48 @@ Description: inference of time stamped phylogenies and ancestral reconstruction
   * inference of GTR models
   * rerooting to obtain best root-to-tip regression
   * auto-correlated relaxed molecular clock (with normal prior)
+ .
+ This package provides the Python 2 module.
+
+Package: python3-treetime
+Architecture: all
+Section: python
+Depends: ${python3:Depends},
+         ${misc:Depends},
+         python3-numpy,
+         python3-scipy,
+         python3-biopython,
+         python3-pandas
+Description: inference of time stamped phylogenies and ancestral reconstruction (Python 3)
+ TreeTime provides routines for ancestral sequence reconstruction and the
+ maximum likelihoo inference of molecular-clock phylogenies, i.e., a tree
+ where all branches are scaled such that the locations of terminal nodes
+ correspond to their sampling times and internal nodes are placed at the
+ most likely time of divergence.
+ .
+ TreeTime aims at striking a compromise between sophisticated
+ probabilistic models of evolution and fast heuristics. It implements GTR
+ models of ancestral inference and branch length optimization, but takes
+ the tree topology as given. To optimize the likelihood of time-scaled
+ phylogenies, treetime uses an iterative approach that first infers
+ ancestral sequences given the branch length of the tree, then optimizes
+ the positions of unconstraine d nodes on the time axis, and then repeats
+ this cycle. The only topology optimization are (optional) resolution of
+ polytomies in a way that is most (approximately) consistent with the
+ sampling time constraints on the tree. The package is designed to be
+ used as a stand-alone tool or as a library used in larger phylogenetic
+ analysis workflows.
+ .
+ Features
+  * ancestral sequence reconstruction (marginal and joint maximum
+    likelihood)
+  * molecular clock tree inference (marginal and joint maximum
+    likelihood)
+  * inference of GTR models
+  * rerooting to obtain best root-to-tip regression
+  * auto-correlated relaxed molecular clock (with normal prior)
+ .
+ This package provides the Python 3 module.
 
 Package: python-treetime-examples
 Architecture: all

debian/python-treetime.links

-usr/share/treetime/ancestral_reconstruction.py	usr/bin/ancestral_reconstruction
-usr/share/treetime/temporal_signal.py		usr/bin/temporal_signal
-usr/share/treetime/timetree_inference.py	usr/bin/timetree_inference

debian/python-treetime.install → debian/python3-treetime.install

 *.py	usr/share/treetime
-usr/lib	usr

debian/rules

@@ -4,8 +4,10 @@
 
 export PYBUILD_NAME = treetime
 
+EXECDIR=debian/python3-$(PYBUILD_NAME)/usr/share/treetime
+
 %:
-	dh $@ --with python2
+	dh $@ --with python2,python3 --buildsystem=pybuild
 
 override_dh_install:
 	dh_install
@@ -14,3 +16,9 @@ override_dh_install:
 	## enable dh_python finding dependency
 	##find debian/*/usr/share/treetime -name "*.py" -exec sed -i 's:#!/usr/bin/env python:#!/usr/bin/python:' \{\} \;
 	find debian/*/usr/lib/*/dist-packages/treetime -name "*.py" -exec sed -i 's:/usr/local/bin/python:/usr/bin/python:' \{\} \;
+	mkdir -p $(EXECDIR)
+	mv debian/python-$(PYBUILD_NAME)/usr/bin/* $(EXECDIR)
+	for py in `find debian/python3-treetime -name "*.py"` ; do \
+	    dh_link -p python3-$(PYBUILD_NAME)	usr/share/treetime/`basename $${py}`	usr/bin/`basename $${py} .py` ; \
+	    sed -i '1s+#!/usr/bin/python+&3+' $${py} ; \
+	done

examples/ancestral_inference.py

@@ -11,7 +11,8 @@ if __name__ == '__main__':
     # load data
     base_name = 'data/H3N2_NA_allyears_NA.20'
     T = Phylo.read(base_name+".nwk", "newick")
-    T.root_with_outgroup(T.find_clades(lambda x:x.name.startswith('A/Scot')).next())
+    OG = list(T.find_clades(lambda x:x.name.startswith('A/Scot')))[0]
+    T.root_with_outgroup(OG)
 
     # instantiate treetime
     myTree = TreeAnc(gtr='Jukes-Cantor', tree = T, aln = base_name+'.fasta', verbose = 0)

examples/basic_clock_tree_diagnostics.py

@@ -68,7 +68,7 @@ if __name__ == '__main__':
     x = np.linspace(-0.1,0.05,1000)+ myTree.tree.root.time_before_present
     Phylo.draw(tree, axes=axs[0], show_confidence=False)
     offset = myTree.tree.root.time_before_present + myTree.tree.root.branch_length
-    cols = sns.color_palette()
+    cols =  ['r', 'g', 'c', 'b', 'm', 'y'] #sns.color_palette()
     depth = myTree.tree.depths()
     for ni,node in enumerate(myTree.tree.find_clades()):
         if (not node.is_terminal()):

examples/ebola.py

@@ -29,11 +29,11 @@ if __name__ == '__main__':
     base_name = 'data/ebola'
     dates = read_dates(base_name)
     # instantiate treetime
-    ebola = TreeTime(gtr='Jukes-Cantor', tree = base_name+'.nwk',
+    ebola = TreeTime(gtr='Jukes-Cantor', tree = base_name+'.nwk', precision=1,
                         aln = base_name+'.fasta', verbose = 4, dates = dates)
 
     # infer an ebola time tree while rerooting and resolving polytomies
-    ebola.run(root='best', relaxed_clock=False, max_iter=2,
+    ebola.run(root='best', relaxed_clock=False, max_iter=2, branch_length_mode='input',
               resolve_polytomies=True, Tc='skyline', time_marginal="assign")
 
     # scatter root to tip divergence vs sampling date

examples/rerooting_and_timetrees.py

@@ -58,7 +58,7 @@ if __name__ == '__main__':
     format_axes(fig, axs)
 
     # rerooting can be done along with the tree time inference
-    tt.run(root="best")
+    tt.run(root="best", branch_length_mode='input')
     # if more complicated models (relaxed clocks, coalescent models) are to be used
     # or you want to resolve polytomies, treetime needs to be run for
     # several iterations, for example as