1
2
3
4
5 """Command-line wrapper for the tree inference program RAxML.
6
7 Derived from the help page for RAxML version 7.3 by Alexandros Stamatakis, but
8 should work for any version 7.X (and probably earlier for most options).
9 """
10 from __future__ import print_function
11 from Bio._py3k import basestring
12
13 from Bio.Application import _Option, _Switch, AbstractCommandline
14
15
17 """Command-line wrapper for the tree inference program RAxML.
18
19 The required parameters are 'sequences' (-s), 'model' (-m) and 'name' (-n).
20 The parameter 'parsimony_seed' (-p) must also be set for RAxML, but if you
21 do not specify it, this wrapper will set the seed to 10000 for you.
22
23 Example:
24
25 >>> from Bio.Phylo.Applications import RaxmlCommandline
26 >>> raxml_cline = RaxmlCommandline(sequences="Tests/Phylip/interlaced2.phy",
27 ... model="PROTCATWAG", name="interlaced2")
28 >>> print(raxml_cline)
29 raxmlHPC -m PROTCATWAG -n interlaced2 -p 10000 -s Tests/Phylip/interlaced2.phy
30
31 You would typically run the command line with raxml_cline() or via
32 the Python subprocess module, as described in the Biopython tutorial.
33
34 Citation:
35
36 Stamatakis A.
37 RAxML-VI-HPC: Maximum Likelihood-based Phylogenetic Analyses with
38 Thousands of Taxa and Mixed Models.
39 Bioinformatics 2006, 22(21):2688-2690.
40
41 Homepage: http://sco.h-its.org/exelixis/software.html
42 """
43
44 - def __init__(self, cmd='raxmlHPC', **kwargs):
45 self.parameters = [
46 _Option(['-a', 'weight_filename'],
47 "Name of a column weight file to assign individual weights "
48 "to each column of the alignment. Those weights must be "
49 "integers separated by any type and number of whitespaces "
50 "within a separate file.",
51 filename=True,
52 equate=False,
53 ),
54
55 _Option(['-b', 'bootstrap_seed'],
56 "Random seed for bootstrapping.",
57 equate=False,
58 ),
59
60 _Option(['-c', 'num_categories'],
61 "Number of distinct rate categories for RAxML when "
62 "evolution model is set to GTRCAT or GTRMIX."
63 "Individual per-site rates are categorized into this "
64 "many rate categories to accelerate computations. "
65 "Default: 25.",
66 equate=False,
67 ),
68
69 _Switch(['-d', 'random_starting_tree'],
70 "Start ML optimization from random starting tree."
71 ),
72
73 _Option(['-e', 'epsilon'],
74 "Set model optimization precision in log likelihood units "
75 "for final optimization of tree topology under MIX/MIXI "
76 "or GAMMA/GAMMAI."
77 "Default: 0.1 for models not using proportion of "
78 "invariant sites estimate; 0.001 for models using "
79 "proportion of invariant sites estimate.",
80 equate=False,
81 ),
82
83 _Option(['-E', 'exclude_filename'],
84 "An exclude file name, containing a specification of "
85 "alignment positions you wish to exclude. Format is "
86 "similar to Nexus, the file shall contain entries like "
87 "'100-200 300-400'; to exclude a single column write, "
88 "e.g., '100-100'. If you use a mixed model, an "
89 "appropriately adapted model file will be written.",
90 filename=True,
91 equate=False,
92 ),
93
94 _Option(['-f', 'algorithm'],
95 """Select algorithm:
96
97 a: Rapid Bootstrap analysis and search for best-scoring ML
98 tree in one program run.
99 b: Draw bipartition information on a tree provided with '-t'
100 based on multiple trees (e.g. form a bootstrap) in a file
101 specifed by '-z'.
102 c: Check if the alignment can be properly read by RAxML.
103 d: New rapid hill-climbing (DEFAULT).
104 e: Optimize model+branch lengths for given input tree under
105 GAMMA/GAMMAI only.
106 g: Compute per site log Likelihoods for one ore more trees
107 passed via '-z' and write them to a file that can be read
108 by CONSEL.
109 h: Compute log likelihood test (SH-test) between best tree
110 passed via '-t' and a bunch of other trees passed via '-z'.
111 i: Perform a really thorough bootstrap, refinement of final
112 bootstrap tree under GAMMA and a more exhaustive algorithm.
113 j: Generate a bunch of bootstrapped alignment files from an
114 original alignemnt file.
115 m: Compare bipartitions between two bunches of trees passed
116 via '-t' and '-z' respectively. This will return the
117 Pearson correlation between all bipartitions found in the
118 two tree files. A file called
119 RAxML_bipartitionFrequencies.outputFileName will be
120 printed that contains the pair-wise bipartition
121 frequencies of the two sets.
122 n: Compute the log likelihood score of all trees contained
123 in a tree file provided by '-z' under GAMMA or
124 GAMMA+P-Invar.
125 o: Old and slower rapid hill-climbing.
126 p: Perform pure stepwise MP addition of new sequences to an
127 incomplete starting tree.
128 s: Split up a multi-gene partitioned alignment into the
129 respective subalignments.
130 t: Do randomized tree searches on one fixed starting tree.
131 w: Compute ELW test on a bunch of trees passed via '-z'.
132 x: Compute pair-wise ML distances, ML model parameters will
133 be estimated on an MP starting tree or a user-defined
134 tree passed via '-t', only allowed for GAMMA-based models
135 of rate heterogeneity.""",
136 checker_function=(lambda x:
137 isinstance(x, basestring) and len(x) == 1),
138 equate=False,
139 ),
140
141 _Option(['-g', 'grouping_constraint'],
142 "File name of a multifurcating constraint tree. "
143 "this tree does not need to be comprehensive, i.e. "
144 "contain all taxa.",
145 filename=True,
146 equate=False,
147 ),
148
149 _Option(['-i', 'rearrangements'],
150 "Initial rearrangement setting for the subsequent "
151 "application of topological changes phase.",
152 equate=False,
153 ),
154
155 _Switch(['-j', 'checkpoints'],
156 "Write checkpoints (intermediate tree topologies)."
157 ),
158
159 _Switch(['-k', 'bootstrap_branch_lengths'],
160 "Print bootstrapped trees with branch lengths. "
161 "The bootstraps will run a bit longer, because model "
162 "parameters will be optimized at the end of each run. "
163 "Use with CATMIX/PROTMIX or GAMMA/GAMMAI."
164 ),
165
166 _Option(['-l', 'cluster_threshold'],
167 "Threshold for sequence similarity clustering. "
168 "RAxML will then print out an alignment to a file "
169 "called sequenceFileName.reducedBy.threshold that "
170 "only contains sequences <= the specified threshold "
171 "that must be between 0.0 and 1.0. RAxML uses the "
172 "QT-clustering algorithm to perform this task. "
173 "In addition, a file called "
174 "RAxML_reducedList.outputFileName will be written "
175 "that contains clustering information.",
176 equate=False,
177 ),
178
179 _Option(['-L', 'cluster_threshold_fast'],
180 "Same functionality as '-l', but uses a less "
181 "exhaustive and thus faster clustering algorithm. "
182 "This is intended for very large datasets with more "
183 "than 20,000-30,000 sequences.",
184 equate=False,
185 ),
186
187 _Option(['-m', 'model'],
188 """Model of Nucleotide or Amino Acid Substitution:
189
190 NUCLEOTIDES:
191
192 GTRCAT : GTR + Optimization of substitution rates + Optimization of site-specific
193 evolutionary rates which are categorized into numberOfCategories distinct
194 rate categories for greater computational efficiency
195 if you do a multiple analysis with '-#' or '-N' but without bootstrapping the program
196 will use GTRMIX instead
197 GTRGAMMA : GTR + Optimization of substitution rates + GAMMA model of rate
198 heterogeneity (alpha parameter will be estimated)
199 GTRMIX : Inference of the tree under GTRCAT
200 and thereafter evaluation of the final tree topology under GTRGAMMA
201 GTRCAT_GAMMA : Inference of the tree with site-specific evolutionary rates.
202 However, here rates are categorized using the 4 discrete GAMMA rates.
203 Evaluation of the final tree topology under GTRGAMMA
204 GTRGAMMAI : Same as GTRGAMMA, but with estimate of proportion of invariable sites
205 GTRMIXI : Same as GTRMIX, but with estimate of proportion of invariable sites
206 GTRCAT_GAMMAI : Same as GTRCAT_GAMMA, but with estimate of proportion of invariable sites
207
208 AMINO ACIDS:
209
210 PROTCATmatrixName[F] : specified AA matrix + Optimization of substitution rates + Optimization of site-specific
211 evolutionary rates which are categorized into numberOfCategories distinct
212 rate categories for greater computational efficiency
213 if you do a multiple analysis with '-#' or '-N' but without bootstrapping the program
214 will use PROTMIX... instead
215 PROTGAMMAmatrixName[F] : specified AA matrix + Optimization of substitution rates + GAMMA model of rate
216 heterogeneity (alpha parameter will be estimated)
217 PROTMIXmatrixName[F] : Inference of the tree under specified AA matrix + CAT
218 and thereafter evaluation of the final tree topology under specified AA matrix + GAMMA
219 PROTCAT_GAMMAmatrixName[F] : Inference of the tree under specified AA matrix and site-specific evolutionary rates.
220 However, here rates are categorized using the 4 discrete GAMMA rates.
221 Evaluation of the final tree topology under specified AA matrix + GAMMA
222 PROTGAMMAImatrixName[F] : Same as PROTGAMMAmatrixName[F], but with estimate of proportion of invariable sites
223 PROTMIXImatrixName[F] : Same as PROTMIXmatrixName[F], but with estimate of proportion of invariable sites
224 PROTCAT_GAMMAImatrixName[F] : Same as PROTCAT_GAMMAmatrixName[F], but with estimate of proportion of invariable sites
225
226 Available AA substitution models: DAYHOFF, DCMUT, JTT, MTREV, WAG, RTREV, CPREV, VT, BLOSUM62, MTMAM, GTR
227 With the optional 'F' appendix you can specify if you want to use empirical base frequencies
228 Please not that for mixed models you can in addition specify the per-gene AA model in
229 the mixed model file (see manual for details)
230 """,
231 equate=False,
232 ),
233
234 _Switch(['-M', 'partition_branch_lengths'],
235 "Switch on estimation of individual per-partition "
236 "branch lengths. Only has effect when used in "
237 "combination with 'partition_filename' ('-q'). "
238 "Branch lengths for individual partitions will be "
239 "printed to separate files. A weighted average of the "
240 "branch lengths is computed by using the respective "
241 "partition lengths. "
242 ),
243
244 _Option(['-n', 'name'],
245 "Name used in the output files.",
246 filename=True,
247 equate=False,
248 ),
249
250 _Option(['-o', 'outgroup'],
251 "Name of a single outgroup or a comma-separated list "
252 "of outgroups, eg '-o Rat' or '-o Rat,Mouse'. In case "
253 "that multiple outgroups are not monophyletic the "
254 "first name in the list will be selected as outgroup. "
255 "Don't leave spaces between taxon names!",
256 checker_function=lambda x: len(x.split()) == 1,
257 equate=False,
258 ),
259
260 _Option(['-q', 'partition_filename'],
261 "File name containing the assignment of models to "
262 "alignment partitions for multiple models of "
263 "substitution. For the syntax of this file please "
264 "consult the RAxML manual.",
265 filename=True,
266 equate=False,
267 ),
268
269 _Option(['-p', 'parsimony_seed'],
270 "Random number seed for the parsimony inferences. "
271 "This allows you to reproduce your results and will "
272 "help developers debug the program. This option HAS "
273 "NO EFFECT in the parallel MPI version.",
274 equate=False,
275 ),
276
277 _Option(['-P', 'protein_model'],
278 "File name of a user-defined AA (Protein) substitution "
279 "model. This file must contain 420 entries, the first "
280 "400 being the AA substitution rates (this must be a "
281 "symmetric matrix) and the last 20 are the empirical "
282 "base frequencies.",
283 filename=True,
284 equate=False,
285 ),
286
287 _Option(['-r', 'binary_constraint'],
288 "File name of a binary constraint tree. "
289 "This tree does not need to be comprehensive, i.e. "
290 "contain all taxa.",
291 filename=True,
292 equate=False,
293 ),
294
295 _Option(['-s', 'sequences'],
296 "Name of the alignment data file, in PHYLIP format.",
297 filename=True,
298 equate=False,
299 ),
300
301 _Option(['-t', 'starting_tree'],
302 "File name of a user starting tree, in Newick format.",
303 filename=True,
304 equate=False,
305 ),
306
307 _Option(['-T', 'threads'],
308 "Number of threads to run. "
309 "PTHREADS VERSION ONLY! "
310 "Make sure to set this at most the number of CPUs "
311 "you have on your machine, otherwise, there will be "
312 "a huge performance decrease!",
313 equate=False,
314 ),
315
316 _Option(['-u', 'num_bootstrap_searches'],
317 "Number of multiple bootstrap searches per replicate. "
318 "Use this to obtain better ML trees for each "
319 "replicate. Default: 1 ML search per bootstrap "
320 "replicate.",
321 equate=False,
322 ),
323
324 _Switch(['-v', 'version'],
325 "Display version information."
326 ),
327
328 _Option(['-w', 'working_dir'],
329 "Name of the working directory where RAxML will "
330 "write its output files. Default: current directory.",
331 filename=True,
332 equate=False,
333 ),
334
335 _Option(['-x', 'rapid_bootstrap_seed'],
336 "Random seed for rapid bootstrapping.",
337 equate=False,
338 ),
339
340 _Switch(['-y', 'parsimony'],
341 "Only compute a parsimony starting tree, then exit."
342 ),
343
344 _Option(['-z', 'bipartition_filename'],
345 "Name of a file containing multiple trees, e.g. from "
346 "a bootstrap run, that shall be used to draw "
347 "bipartition values onto a tree provided with '-t'. "
348 "It can also be used to compute per-site log "
349 "likelihoods in combination with '-f g', and to read "
350 "a bunch of trees for a couple of other options "
351 "('-f h', '-f m', '-f n').",
352 filename=True,
353 equate=False,
354 ),
355
356 _Option(['-N', '-#', 'num_replicates'],
357 "Number of alternative runs on distinct starting trees. "
358 "In combination with the '-b' option, this will invoke a "
359 "multiple bootstrap analysis. "
360 "DEFAULT: 1 single analysis."
361 "Note that '-N' has been added as an alternative since "
362 "'-#' sometimes caused problems with certain MPI job "
363 "submission systems, since '-#' is often used to start "
364 "comments. ",
365 equate=False,
366 ),
367 ]
368 AbstractCommandline.__init__(self, cmd, **kwargs)
369
370 if not self.parsimony_seed:
371 self.parsimony_seed = 10000
372
373
374 if __name__ == "__main__":
375 from Bio._utils import run_doctest
376 run_doctest()
377