40670a5 | stamatak | 15 April 2015, 14:40:19 UTC | -added flags for computing JC69 and K80 DNA models -added an experimental option for ascertainment bias correction that takes into account missing data | 15 April 2015, 14:40:19 UTC |
74ecbe9 | stamatak | 15 April 2015, 14:39:48 UTC | updated manual | 15 April 2015, 14:39:48 UTC |
440a1f5 | stamatak | 13 April 2015, 08:48:58 UTC | Added Windows executables | 13 April 2015, 08:48:58 UTC |
903de58 | stamatak | 10 April 2015, 13:35:09 UTC | updated version number in readme and release notes | 10 April 2015, 13:35:09 UTC |
411e39b | stamatak | 10 April 2015, 13:33:00 UTC | -added ancestral state computation for LG4X and LG4M models -improved partition parser error reporting -added check for preventing users to specify one of the -f options multiple times -imrpoved error checking in file copy of MPI version | 10 April 2015, 13:33:00 UTC |
cb84ec2 | stamatak | 08 February 2015, 10:42:52 UTC | added expliciit offendimng site printout for invariable sites that are in partitions that are to be analyzed using an ascertainment bias correction | 08 February 2015, 10:42:52 UTC |
53e5f18 | stamatak | 19 January 2015, 12:24:18 UTC | When not specifying the number of threads via -T in the PThreads option, RAxML will now set the number of threads to 2 by default. | 19 January 2015, 12:24:18 UTC |
0c4847f | stamatak | 05 January 2015, 17:34:06 UTC | added version 8.1.15 Windows executables kindly provided by William Gearty | 05 January 2015, 17:34:06 UTC |
24e8e40 | stamatak | 27 December 2014, 20:09:06 UTC | small, unimportant bug fix | 27 December 2014, 20:09:06 UTC |
31501e7 | stamatak | 27 December 2014, 19:10:02 UTC | added new protein subst model stmtREV, see: http://datadryad.org/resource/doi:10.5061/dryad.7b470 | 27 December 2014, 19:10:02 UTC |
a25e3a0 | stamatak | 22 December 2014, 14:58:48 UTC | --flag-check option now also writes output into a file called RAxML_flagCheck | 22 December 2014, 14:58:48 UTC |
52d3f18 | stamatak | 19 December 2014, 13:38:15 UTC | -added some additional error checking for partitions and the EPA algorithm, to prevent users from doing things wrong | 19 December 2014, 13:38:15 UTC |
81a791f | stamatak | 18 December 2014, 18:33:23 UTC | added v 8.1.13 windows executables | 18 December 2014, 18:33:23 UTC |
04ae5d2 | stamatak | 17 December 2014, 08:06:15 UTC | added dll file to new windows executables | 17 December 2014, 08:06:15 UTC |
f72f782 | stamatak | 17 December 2014, 08:04:04 UTC | Added Windows executables for version 8.1.11 | 17 December 2014, 08:04:04 UTC |
9706247 | stamatak | 16 December 2014, 14:08:33 UTC | fixed a bug in lewis correction for asc. bias | 16 December 2014, 14:08:33 UTC |
cd236b5 | stamatak | 12 December 2014, 15:56:29 UTC | fixed the bugs introduced by the bug fix | 12 December 2014, 15:56:29 UTC |
97a901d | stamatak | 12 December 2014, 15:50:29 UTC | fixed a bug in lewis ascertainment bias correction, the likelihoods were not correctly scaled in evaluateGeneric() | 12 December 2014, 15:50:29 UTC |
041e083 | stamatak | 11 December 2014, 15:45:13 UTC | fixed small bug in LG4X model when writeVector is set to TRUE | 11 December 2014, 15:45:13 UTC |
0816615 | stamatak | 10 December 2014, 21:42:42 UTC | fixed some issues with undetermined sites supression | 10 December 2014, 21:42:42 UTC |
13a6f4c | stamatak | 10 December 2014, 17:29:20 UTC | added functionality to automatically remove fully undetermined sites from the alignment | 10 December 2014, 17:29:20 UTC |
7430f22 | Alexis Stamatakis | 08 December 2014, 12:42:48 UTC | Merge pull request #17 from wookietreiber/master use more portabel perl shebangs | 08 December 2014, 12:42:48 UTC |
1ea2418 | Christian Krause | 07 December 2014, 12:09:06 UTC | use more portabel perl shebangs use `/usr/bin/env perl` instead of `/usr/local/bin/perl` as it finds perl in non-hard-coded locations | 07 December 2014, 12:09:06 UTC |
fe7c8a9 | stamatak | 05 December 2014, 18:05:02 UTC | fixed a bug in the parser | 05 December 2014, 18:05:02 UTC |
203f9b8 | stamatak | 05 December 2014, 17:37:49 UTC | fixed some errors in the protein model selection test | 05 December 2014, 17:37:49 UTC |
427762d | stamatak | 05 December 2014, 16:00:43 UTC | update version numbers | 05 December 2014, 16:00:43 UTC |
e7f0ed4 | stamatak | 05 December 2014, 15:43:56 UTC | fixed a problem in the auto prot output | 05 December 2014, 15:43:56 UTC |
36ec4a9 | stamatak | 05 December 2014, 14:19:56 UTC | small fixes for auto protein | 05 December 2014, 14:19:56 UTC |
3e0e1f5 | stamatak | 05 December 2014, 11:00:49 UTC | updated manual | 05 December 2014, 11:00:49 UTC |
f6656d6 | stamatak | 02 December 2014, 10:52:46 UTC | updated manual | 02 December 2014, 10:52:46 UTC |
f3ea3d6 | stamatak | 28 November 2014, 14:06:22 UTC | -added two additional ways for correcting ascertainment bias. Both new methods can be used to correct for the absence of invariable sites, when the number of invariable sites of the dataset is known. | 28 November 2014, 14:06:22 UTC |
03479cc | stamatak | 28 November 2014, 14:05:30 UTC | updated manual | 28 November 2014, 14:05:30 UTC |
5a695ce | stamatak | 26 November 2014, 07:29:06 UTC | fixed a bug in the Pthreads version of the evolutionary placement algorithm (EPA) that occured under the CAT model. | 26 November 2014, 07:29:06 UTC |
7cc688d | stamatak | 21 November 2014, 15:53:06 UTC | the --mesquite option will now generate some additional output files | 21 November 2014, 15:53:06 UTC |
ceb4dde | stamatak | 21 November 2014, 13:14:42 UTC | some modifications for mesquite-compatible output | 21 November 2014, 13:14:42 UTC |
2e3eda2 | stamatak | 21 November 2014, 11:48:18 UTC | added some experimental stuff for bastien to play with | 21 November 2014, 11:48:18 UTC |
bc92d59 | stamatak | 17 November 2014, 20:07:55 UTC | updated manual | 17 November 2014, 20:07:55 UTC |
92be274 | stamatak | 17 November 2014, 19:56:34 UTC | improved error checking for safer use of CAT model with binary model files. | 17 November 2014, 19:56:34 UTC |
10ccb29 | stamatak | 17 November 2014, 14:47:44 UTC | 1. added some error checking for making sure that binary model files that are being read and the current command line use the same model of rate heterogeneity 2. corrected error output text in the parser | 17 November 2014, 14:47:44 UTC |
90a7af0 | stamatak | 17 November 2014, 14:23:32 UTC | improved parser error handling | 17 November 2014, 14:23:32 UTC |
0bfa271 | stamatak | 17 November 2014, 12:51:52 UTC | added GPL license | 17 November 2014, 12:51:52 UTC |
529e55c | stamatak | 17 November 2014, 11:29:05 UTC | some code cleanup for removing warnings generated by clang | 17 November 2014, 11:29:05 UTC |
110f3b8 | stamatak | 17 November 2014, 10:58:25 UTC | updated Makefiles and version number | 17 November 2014, 10:58:25 UTC |
14e313c | stamatak | 17 November 2014, 10:32:56 UTC | fixed a bug in the parser that woould enable ASC bias correction for protein susbst models by default | 17 November 2014, 10:32:56 UTC |
594dda7 | stamatak | 17 November 2014, 10:05:03 UTC | -corrected a bug that caused the bootstrap searches to fail in the Pthreads version of the code. This bug was associated with the additional optimization of the likelihood functions based on Andre Aberer's idea. | 17 November 2014, 10:05:03 UTC |
e8634fe | stamatak | 04 November 2014, 11:17:58 UTC | fixed a bug ine the PTHREADS version of the BFGS optimization | 04 November 2014, 11:17:58 UTC |
a0dc884 | stamatak | 04 November 2014, 07:19:13 UTC | fixed a bug that Alexey detected for the Pthreads version of the subtree EPA algorithm | 04 November 2014, 07:19:13 UTC |
25149d0 | stamatak | 04 November 2014, 07:02:07 UTC | fixed small incompatible types issue that occured while trying to compile for windows | 04 November 2014, 07:02:07 UTC |
df2f584 | stamatak | 04 November 2014, 06:50:35 UTC | fixed bug in BFGS optimizationy | 04 November 2014, 06:50:35 UTC |
051fd11 | stamatak | 25 August 2014, 05:32:28 UTC | fixed a broken link in the README file | 25 August 2014, 05:32:28 UTC |
fda59b8 | stamatak | 14 August 2014, 06:21:33 UTC | added include sys/types.h to axml.h | 14 August 2014, 06:21:33 UTC |
fbbca07 | stamatak | 13 August 2014, 13:00:56 UTC | changed output tree format for subtree placements for Alexey | 13 August 2014, 13:00:56 UTC |
f27a1c9 | stamatak | 05 August 2014, 17:56:06 UTC | added an experimental option for Alexey | 05 August 2014, 17:56:06 UTC |
2aef615 | stamatak | 31 July 2014, 12:30:15 UTC | 1. some code cleanup 2. Finished integration of BFGS method for simultaneous optimization of GTR rate parameters. It will only be invoked on single-partition DNA datasets, because a version for several partitions requires substantially more work, since paremeters for all partitions need to be changed simultaneously following the approach described here: http://sco.h-its.org/exelixis/pubs/ICPP2009.pdf Performance-wise I observed a speed improvement of up to 32% for optimizing model params & br-lens on a fixed single-gene tree with 2554 taxa 3. To disable BFGS optimization in case of bugs I added a new flag --no-bfgs to disable bfgs optimization. | 31 July 2014, 12:30:15 UTC |
c9c0e16 | stamatak | 31 July 2014, 12:27:16 UTC | updated manual | 31 July 2014, 12:27:16 UTC |
da5f026 | stamatak | 31 July 2014, 07:39:46 UTC | some code cleanup | 31 July 2014, 07:39:46 UTC |
d5d01df | stamatak | 30 July 2014, 11:13:48 UTC | 1. updated the manual 2. commented and improved some of Kassians code for TC/IC on partial gene trees 3. Integrated the optimization of the DNA+GAMMA models, that is, unecessary precalculations of the P-Matrix are omitted using Andre's approach (yields between 3-5%, performance benefits on smaller partitions will be larger) 4. Added three new options: --mesquite to print outputfiles for interacting with mesquite --silent to avoid printing out info about identical sequences and fully undetermined patterns in the MSA --no-seq-check deactivates checking of the MSA for duplicate seqs and undetermined patterns, may save quite some time because the run time for this is O(n^2 + m^2) where n is the number of taxa and m the number of patterns 5. Changed command line parsing back to the FNU function that can now also accept long options like --mesquite etc. | 30 July 2014, 11:22:15 UTC |
0df460f | stamatak | 23 July 2014, 13:01:01 UTC | Added experimental code for corrected TC/IC calculations on partial gene trees. The code is not activated yet, only there for testing. | 23 July 2014, 13:01:01 UTC |
d8630ca | stamatak | 27 June 2014, 12:38:51 UTC | updated manual | 27 June 2014, 12:38:51 UTC |
0d743b8 | stamatak | 27 June 2014, 09:39:11 UTC | updated manual | 27 June 2014, 09:39:11 UTC |
dae223e | stamatak | 25 June 2014, 18:46:17 UTC | -added code to prevent raxml from generating a seg-fault and returning a useful error message instead when trying to parse taxon labels in a tree that consist of an empty string | 25 June 2014, 18:46:17 UTC |
94067d7 | stamatak | 16 June 2014, 13:52:44 UTC | updated manual | 16 June 2014, 13:52:44 UTC |
b54f5bd | stamatak | 16 June 2014, 13:30:58 UTC | updated version number in README and Release-Notes | 16 June 2014, 13:30:58 UTC |
9f9ad49 | stamatak | 16 June 2014, 13:26:12 UTC | -added a simple heterotachous model to RAxML, for details, see manual | 16 June 2014, 13:26:12 UTC |
c4e1307 | stamatak | 09 June 2014, 20:17:46 UTC | updated manual | 09 June 2014, 20:17:46 UTC |
f4ce699 | stamatak | 06 June 2014, 14:07:35 UTC | added a more elaborate function for imputing missing sequence data. Needs to be activated by uncommenting: #define _SUBTREE_PREDICTION in axml.c | 06 June 2014, 14:07:35 UTC |
934296f | stamatak | 23 May 2014, 09:26:50 UTC | updated manual | 23 May 2014, 09:26:50 UTC |
03e580a | stamatak | 06 May 2014, 16:30:03 UTC | fixed an assertion that shall not fail when bootstrapping using a wieght file | 06 May 2014, 16:30:03 UTC |
82fa3b5 | stamatak | 06 May 2014, 15:38:12 UTC | added windows executables for v 8.0.20 kindly provided by Ingo Michalak :-) | 06 May 2014, 15:38:12 UTC |
b7b849d | stamatak | 06 May 2014, 15:29:37 UTC | fixed a bug in the tree string processing | 06 May 2014, 15:29:37 UTC |
6c39ac9 | stamatak | 18 April 2014, 06:31:58 UTC | updated manual | 18 April 2014, 06:31:58 UTC |
a6506ee | stamatak | 13 April 2014, 08:48:49 UTC | fixed minor type casting issues and minor bug in error detection output message | 13 April 2014, 08:48:49 UTC |
c343283 | stamatak | 28 March 2014, 09:43:16 UTC | 1. fixed a bug in the LG4X protein substitution model 2. fixed a bug for the ascertainment bias correction calculation with partitioned datasets | 28 March 2014, 09:43:16 UTC |
7df9705 | stamatak | 27 March 2014, 14:06:11 UTC | uncommented the wrong debug flag commented the correct debug flag | 27 March 2014, 14:06:11 UTC |
dcbe679 | stamatak | 27 March 2014, 14:03:02 UTC | commented a debug flag | 27 March 2014, 14:03:02 UTC |
4a1e9e2 | stamatak | 27 March 2014, 14:00:59 UTC | -fixed a bug in the error checking function -cleaned up the code a bit | 27 March 2014, 14:00:59 UTC |
1bf43c6 | stamatak | 26 March 2014, 10:26:27 UTC | -fixed a typo in a warning | 26 March 2014, 10:26:27 UTC |
334c163 | stamatak | 25 March 2014, 17:31:05 UTC | updated windows executables | 25 March 2014, 17:31:05 UTC |
a693459 | stamatak | 25 March 2014, 16:19:31 UTC | Fixed bugs in: 1. The PThreads version of the evolutionary placement algorithm 2. The calculation of SH-like support values | 25 March 2014, 16:19:31 UTC |
38cb4d1 | stamatak | 25 March 2014, 11:29:01 UTC | Fixed Makefiles for Macintosh using clang compiler. | 25 March 2014, 11:29:01 UTC |
623387c | stamatak | 22 March 2014, 07:23:39 UTC | added readme to windows executables directory | 22 March 2014, 07:23:39 UTC |
b526883 | stamatak | 22 March 2014, 07:22:17 UTC | updated windows executables | 22 March 2014, 07:22:17 UTC |
c18cc11 | stamatak | 21 March 2014, 13:04:07 UTC | updated manual | 21 March 2014, 13:04:07 UTC |
5361156 | stamatak | 19 March 2014, 17:09:54 UTC | fixed some typos in the printouts, made the makefiles debian compliant | 19 March 2014, 17:09:54 UTC |
ef51969 | stamatak | 16 March 2014, 17:36:50 UTC | added directory with some windows executables compiled by Ingo Michalak. | 16 March 2014, 17:36:50 UTC |
e12b993 | stamatak | 13 March 2014, 17:16:30 UTC | removed some unnecessary source files. | 13 March 2014, 17:16:30 UTC |
2221aa2 | stamatak | 12 March 2014, 17:22:16 UTC | added some source code documentation | 12 March 2014, 17:22:16 UTC |
2fb2ec2 | stamatak | 11 March 2014, 19:34:50 UTC | fixed a bug in the fast tree search algorithm (-f E) option. | 11 March 2014, 19:34:50 UTC |
6745b06 | stamatak | 11 March 2014, 18:58:19 UTC | updated manual | 11 March 2014, 18:58:19 UTC |
a869b6b | stamatak | 11 March 2014, 16:40:26 UTC | updated version numbers | 11 March 2014, 16:40:26 UTC |
4f1b8ce | stamatak | 11 March 2014, 16:38:40 UTC | bug fixes in fasta file pasring and LG4X model. | 11 March 2014, 16:38:40 UTC |
010c3b4 | stamatak | 11 March 2014, 16:22:10 UTC | updated manual | 11 March 2014, 16:22:10 UTC |
693b12f | stamatak | 06 March 2014, 17:43:40 UTC | improved error reporting in partition file parser. | 06 March 2014, 17:43:40 UTC |
9f2378a | stamatak | 05 March 2014, 15:54:28 UTC | added some additional checks for the base frequency range to prevent potential numerical problems. | 05 March 2014, 15:54:28 UTC |
5b1f6aa | stamatak | 04 March 2014, 16:56:04 UTC | improved the error checking for the LnL tolerance in model parameter optimization routine. The tolerance is not hard-coded but a function of the LnL score now. | 04 March 2014, 16:56:04 UTC |
6e8356e | stamatak | 28 February 2014, 14:41:15 UTC | minor fix such that the -f B option works again, this was broken by the introduction of the ML estimate of base frequencies. | 28 February 2014, 14:41:15 UTC |
da95c3f | stamatak | 28 February 2014, 14:08:20 UTC | updated manual | 28 February 2014, 14:08:20 UTC |
5cc79f6 | stamatak | 28 February 2014, 12:33:05 UTC | Fixed an issue with an assertion for the likelihood calculations that failed. The assertion was placed incorrectly. | 28 February 2014, 12:33:05 UTC |
145f77e | stamatak | 28 February 2014, 08:01:19 UTC | updated manual | 28 February 2014, 08:01:19 UTC |
d127945 | stamatak | 27 February 2014, 12:26:38 UTC | made the error diagnosis for tree file parsing more verbose and understandable, RAxML will now also exit with an error when instead of a tree file you have specified an alignment file | 27 February 2014, 12:26:38 UTC |