GARLI 2.10 on XSEDE 2.10 Genetic Algorithm for Rapid Likelihood Inference - run on XSEDE. Derrick J. Zwickl (zwickl@nescent.org) Zwickl, D. J., 2006. Genetic algorithm approaches for the phylogenetic analysis of large biological sequence datasets under the maximum likelihood criterion. Ph.D. dissertation, The University of Texas at Austin. http://garli.nescent.org Phylogeny / Alignment garli2_tgb garli perl "garli_2.1_expanse" 0 infile Alignment file (non-interleaved phylip or nexus format) (s) 0 infile number_of_runssearchrep1 1 perl $bootstrapreps < 1 && $searchreps_value > 0 perl "$searchreps_value" number_of_runsbootrep 1 perl $bootstrapreps > 0 perl "$bootstrapreps" jobtype 1 scheduler.conf perl "jobtype=mpi\\n" this_is_a_workflow scheduler.conf perl "workflow_type=simple\\n" 1 searchrep_mpi_processes_nodeslt32 1 scheduler.conf perl "nodes=1 \\n" collectboottrees 98 perl $bootstrapreps > 1 && $searchreps_value == 1 perl "&& /expanse/projects/ngbt/home/cipres/ngbw/contrib/tools/bin/collectBootTrees.sh" findbesttrees 99 perl $bootstrapreps < 1 && $searchreps_value > 1 perl "&& python3 /expanse/projects/ngbt/home/cipres/ngbw/contrib/tools/bin/post_garli.py" findbesttrees2 99 perl $bootstrapreps > 1 && $searchreps_value > 1 perl "&& python3 /expanse/projects/ngbt/home/cipres/ngbw/contrib/tools/bin/post_garli.py" searchrep_mpi_processes 1 scheduler.conf perl !$userconffilethere && $bootstrapreps < 1 && $searchreps_value > 0 && $searchreps_value < 50 && !$more_memory perl "mpi_processes=$searchreps_value\\n" . "mem=" . ($searchreps_value * 2) . "G\\n" . "node_exclusive=0\\n" searchrep_mpi_processesb 1 scheduler.conf perl !$userconffilethere && $bootstrapreps < 1 && $searchreps_value > 0 && $searchreps_value < 50 && $more_memory perl "mpi_processes=$searchreps_value\\n" . "mem=" . ($searchreps_value * 4) . "G\\n" . "node_exclusive=0\\n" searchrep_mpi_processes_gt49 1 scheduler.conf perl !$userconffilethere && $bootstrapreps < 1 && $searchreps_value > 49 && !$more_memory perl "mpi_processes=50\\n" . "mem=100G\\n" . "node_exclusive=0\\n" 49 procs with more memory --> searchrep_mpi_processes_gt49b 1 scheduler.conf perl !$userconffilethere && $bootstrapreps < 1 && $searchreps_value > 49 && $more_memory perl "mpi_processes=58\\n" . "mem=243G\\n" . "node_exclusive=1\\n" mpi_processes_1-49 1 scheduler.conf perl !$userconffilethere && $bootstrapreps > 0 && $bootstrapreps < 50 && !$more_memory perl "mpi_processes=$bootstrapreps\\n" . "mem=" . ($bootstrapreps * 2) . "G\\n" . "node_exclusive=0\\n" mpi_processes_1-49b 1 scheduler.conf perl !$userconffilethere && $bootstrapreps > 0 && $bootstrapreps < 50 && $more_memory perl "mpi_processes=$bootstrapreps\\n" . "mem=" . ($bootstrapreps * 4) . "G\\n" . "node_exclusive=0\\n" mpi_processes_gt49 1 scheduler.conf perl !$userconffilethere && $bootstrapreps > 49 && !$more_memory perl "mpi_processes=50\\n" . "mem=100G\\n" . "node_exclusive=0\\n" mpi_processes_gt49b 1 scheduler.conf perl !$userconffilethere && $bootstrapreps > 49 && $more_memory perl "mpi_processes=50\\n" . "mem=243G\\n" . "node_exclusive=1\\n" searchrep_mpi_processes_userconf 1 scheduler.conf perl $userconffilethere && $bootstrapreps < 1 && $searchreps_value > 0 && ($set_divvalue*$searchreps_value*2) < 200 perl "mpi_processes=$searchreps_value\\n" . "mem=" . ($set_divvalue*$searchreps_value*2) . "G\\n" . "node_exclusive=0\\n" searchrep_mpi_processes_gt50b 1 scheduler.conf perl $userconffilethere && $bootstrapreps < 1 && ($set_divvalue*$searchreps_value*2) > 199 perl "mpi_processes=50\\n" . "mem=243G\\n" . "node_exclusive=1\\n" mpi_processes_1-49_userconf 1 scheduler.conf perl $userconffilethere && $bootstrapreps > 0 && ($set_divvalue*$bootstrapreps*2) < 200 perl "mpi_processes=$bootstrapreps\\n" . "mem=" . ($set_divvalue*$bootstrapreps*2) . "G\\n" . "node_exclusive=0\\n" mpi_processes_1-49_userconfb 1 scheduler.conf perl $userconffilethere && $bootstrapreps > 0 && ($set_divvalue*$bootstrapreps*2) > 199 perl "mpi_processes=$bootstrapreps\\n" . "mem=243G\\n" . "node_exclusive=1\\n" runtime 1 scheduler.conf Maximum Hours to Run (click here for help setting this correctly) 0.5 Maximum Hours to Run must be between 0.1 - 168.0. perl $runtime < 0.1 || $runtime > 168.0 The job will run on $searchreps_value processors as configured. If it runs for the entire configured time, it will consume $searchreps_value x $runtime cpu hours perl !$userconffilethere && $bootstrapreps < 1 && $searchreps_value > 0 && $searchreps_value < 50 && !$more_memory The job will run on 2 x $searchreps_value processors as configured. If it runs for the entire configured time, it will consume 2 x $searchreps_value x $runtime cpu hours perl !$userconffilethere && $bootstrapreps < 1 && $searchreps_value > 0 && $searchreps_value < 50 && $more_memory The job will run on $searchreps_value x $set_divvalue processors as configured. If it runs for the entire configured time, it will consume $searchreps_value x $set_divvalue x $runtime hours perl $userconffilethere && $bootstrapreps < 1 && $searchreps_value > 0 && (($set_divvalue * $searchreps_value * 2) < 200) The job will run on 128 processors as configured. If it runs for the entire configured time, it will consume 128 x $runtime hours perl $userconffilethere && $bootstrapreps < 1 && $searchreps_value > 0 && (($set_divvalue * $searchreps_value * 2) > 200) The job will run on $bootstrapreps processors as configured. If it runs for the entire configured time, it will consume $bootstrapreps x $runtime cpu hours perl !$userconffilethere && $bootstrapreps > 0 && $bootstrapreps < 50 && !$more_memory The job will run on 50 processors as configured. If it runs for the entire configured time, it will consume 50 x $runtime cpu hours perl !$userconffilethere && $bootstrapreps > 49 && !$more_memory The job will run on $bootstrapreps processors as configured. If it runs for the entire configured time, it will consume $bootstrapreps x $runtime cpu hours perl !$userconffilethere && $bootstrapreps > 0 && $searchreps_value < 2 && $bootstrapreps < 50 && !$more_memory The job will run on 2 x $bootstrapreps processors as configured. If it runs for the entire configured time, it will consume 2 x $bootstrapreps x $runtime cpu hours perl !$userconffilethere && $bootstrapreps > 0 && $searchreps_value < 2 && $bootstrapreps < 50 && $more_memory The job will run on 50 processors as configured. If it runs for the entire configured time, it will consume 50 x $runtime cpu hours perl !$userconffilethere && $bootstrapreps > 0 && $bootstrapreps < 50 && $more_memory The job will run on 128 processors as configured. If it runs for the entire configured time, it will consume 128 x $runtime cpu hours perl !$userconffilethere && $bootstrapreps > 49 && $more_memory The job will run on $bootstrapreps processors as configured. If it runs for the entire configured time, it will consume $bootstrapreps x $runtime cpu hours perl $userconffilethere && $bootstrapreps > 0 && ($set_divvalue * $bootstrapreps * 2) < 50 The job will run on 50 processors as configured. If it runs for the entire configured time, it will consume 50 x $runtime cpu hours perl $userconffilethere && $bootstrapreps > 0 && ($set_divvalue * $bootstrapreps * 2) >= 50 perl "runhours=$value\\n" Estimate the maximum time your job will need to run. We recommend testimg initially with a < 0.5hr test run because Jobs set for 0.5 h or less depedendably run immediately in the "debug" queue. Once you are sure the configuration is correct, you then increase the time. The reason is that jobs > 0.5 h are submitted to the "normal" queue, where jobs configured for 1 or a few hours times may run sooner than jobs configured for the full 168 hours. user_conffile 1 I would like to provide my own GARLI configuration file (for help, see the GARLI.conf Creator tool) This option allows the user to upload a garli.conf file that has multiple partitions. One can create a garli.conf file with up to 5 partitions using the GARLI.conf Creator tool here, or follow the instructions in the GARLI manual to create your own file. The files created using GARLI.conf Creator can be edited manually .to manage more (unlimited) partitions. userconffilehere 1 The garli.conf file was created using GARLI.conf Creator or the GARLI2_TG interface perl $user_conffile && !$userconffilethere Please specify whether the garli.conf file was created using the CIPRES GARLI.conf Creator tool, or elsewhere perl !$userconffilethere && !$userconffilehere A garli.conf file created by the Garli Conf Creator will be configured correctly for the CIPRES Science Gateway, so no further adjustments will be needed. userconffilethere 1 The garli.conf file was created outside of CIPRES perl $user_conffile && !$userconffilehere Please be sure that in the garli.conf file you uploaded, searchreps=1 (even though this seems counterintuitive; it is essential for the mpi code), then click the box that says I have set the values of searchreps and bootstrapreps correctly perl $user_conffile && $userconffilethere && $bootstrapreps < 1 && $searchreps_value > 1 Please be sure that in the garli.conf file you uploaded, bootstrapreps=1 (even though this seems counterintuitive; it is essential for the mpi code), then click the box that says I have set the values of searchreps and bootstrapreps correctly perl $user_conffile && $userconffilethere && $bootstrapreps > 1 A garli.conf file created without using the Garli Conf Creator may not be configured correctly for the CIPRES Science Gateway. Please make sure your file conforms to the following characteristics: If bootstrap reps=0 please be sure that searchreps=1. If bootstrapreps > 0 please be sure thta in the garli.conf file you uploaded bootstrapreps = 1 ; and searchreps = the number of searchreps you actually want (even though this seems counterintuitive; it is essential for the mpi code. upload_conffile Select the GARLI configuration file perl $user_conffile && ($userconffilethere || $userconffilehere) 1 garli.conf Please select a garli.conf file perl !defined $upload_conffile && ($userconffilethere || $userconffilehere) more_memory I need more memory This option allows the user obtained more memory. This will increase the cost of the run set_divvalue 1 How much memory did you specify in availablememory in your conf file perl $userconffilethere 1 2 3 4 5 Please specify how much memory you requested in your garli.conf file perl $userconffilethere && !defined $set_divvalue userconffileconfirm 1 I have set the values of searchreps and bootstrapreps correctly perl $userconffilethere a_matrix_file Input infile b_command_line commandline COMMANDLINE c_garli_configuration Input parameters garli.conf d_starting_tree Input starting tree/model starting.txt perl defined $streefname_pasted || defined $streefname_userdata e_constraint_file Input topology constraints constraint.txt perl defined $constraintfile_control f_scheduler_configuration Scheduling parameters scheduler.conf ff_scheduler_configuration Scheduling stderr scheduler_stderr.txt g_jobinfo_txt jobinfo_parameters *_JOBINFO.TXT h_start_txt start_txt start.txt hh_done_txt done_txt done.txt i_garli_log00_results *.log00.log j_garli_screenlog_results *.screen.log k_garli_boottree_results *boot.tre l_garli_allboottree_results allBootTrees.tre m_garli_besttree_results *best.tre n_garli_besttree_results all.tre perl $bootstrapreps < 1 o_garli_bestalltree_results *best.all.tre p_garli_treescores_results garli_scores.txt all_results * general_section garli.conf perl !$user_conffile 1 perl "[general]\\n" master_section garli.conf perl !$user_conffile 3 perl "\n\n[master]\\n" datafname garli.conf perl !$user_conffile 2 perl "datafname = infile\\n" 2 ofprefix garli.conf perl !$user_conffile 2 perl "ofprefix = garli_run\\n" 3 availablememory garli.conf perl !$user_conffile 2 perl "availablememory = 2000\\n" configure_partitions Partition Configuration is_partitioned perl !$user_conffile 1 The input is a Nexus file that specifies at least two partitions linkmodels Apply selected model parameters to all partitions 2 perl $is_partitioned && !$user_conffile perl ($value) ? "linkmodels = 1\\n":"" garli.conf Sorry, you cannot apply different rates to each partition unless you upload a garli.conf file perl !$linkmodels && !$user_conffile subsetspecificrates Infer overall rate multipliers for each data partition 2 perl !$user_conffile && $is_partitioned perl ($value) ? "subsetspecificrates = $value\\n" : "" 0 garli.conf The parameter subsetspecificrates means to infer overall rate multipliers for each data subset. This is equivalent to *prset ratepr=variable* in MrBayes So, using this interface you can make the following settings: subsetspecificrates = 0 single model, one set of branch lengths (equivalent to non-partitioned analysis) subsetspecificrates = 1 single model, different subset rates (like site-specific rates model in PAUP*) run_configuration Run Configuration streefname_choose Specify where the starting tree topology comes from (streefname) perl !$user_conffile stepwise random stepwise upload random "streefname = random\\n" stepwise "streefname = stepwise\\n" upload "streefname = starting.txt\\n" garli.conf 2 streefname Specifies where the starting tree topology and/or model parameters will come from. The tree topology may be a completely random topology (constraints will be enforced), a tree provided by the user in a file, or a tree generated by the program using a fast ML stepwise-addition algorithm (see attachmentspertaxon below). The author recommends stepwise over random. This will give it a much better starting point; on large datasets this can both reduce runtimes and improve results.Starting or fixed model parameter values may also be provided in the specified file, with or without a tree topology. Some notes on starting trees/models: Nexus starting trees are allowed. Specified starting trees may have polytomies, which will be arbitrarily resolved before the run begins. Allowed Starting tree formats: Plain newick tree string (with taxon numbers or names, with or without branch lengths) or NEXUS trees block. The trees block can appear in the same file as a NEXUS data or characters block that contains the alignment, but it must specfied a second time in this box. If multiple trees appear in the specified file and multiple search replicates are specified (see searchreps setting), then the first tree is used in the first replicate the second in the second replicate, etc. streefname_userdata perl $streefname_choose ne "random" && $streefname_choose ne "stepwise" garli.conf 2 Choose a Starting Tree (streefname file) starting.txt In order to work, your garli.conf file must contain the one line streefname = starting.txt in the [general] section of garli.conf. No other streefname = lines should be present. perl ($userconffilethere || $userconffilehere) && $streefname_choose eq "upload" Specify the name of the file containing the population starting conditions. Starting model parameters and/or a starting topology may be specified. If both model and topology are specified, the model must come first, and both must appear on the same line of the file. Each model parameter is specified by a letter representing the parameter type, followed by the value or values assigned. Thus r 1.4 3.4 0.55 1.09 4.94 b 0.297 0.185 0.213 0.305 a 0.66 p 0.43 ((((((140:..etc specifies starting values for the rate matrix (in the order AC, AG, AT, CG, CT), base frequencies (in the order A, C, G, T), alpha shape of the gamma rate-heterogeneity distribution and the proportion of invariable sites, and is followed by the starting tree. If starting parameters are not specified, the base frequencies begin at their empirical values, the proportion of invariable sites begins at 20% of the observed proportion of invariants sites, alpha starts at 0.5 and the rate matrix starts at values equivalent to a kappa value of 5.0. If included, the tree specification should appear in Newick format (parenthetical notation), with the taxa represented by either their name or their number in the data matrix (starting at 1). Starting branch lengths on the tree are optional. The sample dataset included with the program comes with an example of a starting model/tree file. As of version 0.95, various models may be specified and various parameter values fixed. Any model parameters that appear in the starting condition file must correspond to the model chosen in the config file. In addition, any parameters specified as fixed in the config file must have their values specified here. If no file is specified, a random start tree and default model parameters will be inferred. attachments_val Specify the number of attachment branches evaluated for each taxon (attachmentspertaxon) perl !$user_conffile && $streefname_choose eq "stepwise" 50 perl "attachmentspertaxon=$value\\n" garli.conf 2 New Param for V 0.96. The number of attachment branches evaluated for each taxon to be added to the tree during the creation of an ML stepwise-addition starting tree. Briefly, stepwise addition is an algorithm used to make a tree, and involves adding taxa in a random order to a growing tree. For each taxon to be added, a number of randomly chosen attachment branches are tried and scored, and then the best scoring one is chosen as the location of that taxon. The attachmentspertaxon setting controls how many attachment points are evaluated for each taxon to be added. A value of one is equivalent to a completely random tree (only one randomly chosen location is evaluated). A value of greater than 2 times the number of taxa in the dataset means that all attachment points will be evaluated for each taxon, and will result in very good starting trees (but may take a while on large datasets). Even fairly small values (less than 10) can result in starting trees that are much, much better than random, but still fairly different from one another. searchreps_value Specify the number of independent search replicates to perform during a program execution.(searchreps) 2 The value for searchreps cannot be less than 1 perl $searchreps_value < 1 You must specify multiple searchreps and/or multiple bootstrapreps to do a parallel run. perl $searchreps_value < 2 && $bootstrapreps < 2 Sorry, you cant do both a regular search and bootstrapping here perl $searchreps_value > 1 && $bootstrapreps > 0 Sorry, the number of searchreps must be no more than 120. perl $searchreps_value > 120 Please enter an integer value for search repetitions perl !defined $searchreps_value The number of independent search replicates to perform during a program execution. You should always either do multiple search replicates or multiple program executions with any dataset to get a feel for whether you are getting consistent results, which suggests that the program is doing a decent job of searching. Note that if this is greater than 1 and you are performing a bootstrap analysis, this is the number of search replicates to be done per bootstrap replicate. That can increase the chance of finding the best tree per bootstrap replicate, but will also increase bootstrap runtimes enormously. When you specify multiple searchreps or multiple bootstraps, we will automatically run garli multiple times in parallel. searchreps_value_no_upload Specify the number of independent search replicates to perform during a program execution.(searchreps) perl !$user_conffile perl ($bootstrapreps > 1) ? "searchreps=$searchreps_value\\n" : "searchreps=1\\n" garli.conf 2 bootstrapreps Bootstrap Repetitions (-bootstrapreps) perl !$inferinternalstateprobs_g 0 Bootstrap Repetitions must be 0 or an integer 2 or greater for parallel runs. perl $value < 0 || $value == 1 Please enter an integer value for Bootstrap Repetitions perl !defined $bootstrapreps The number of bootstrap reps to perform. If the value entered is 0,normal searching will be performed. If a value greater than 0 is entered, normal searching will not be performed. The resulting bootstrap trees (one per rep) will be output to a file named ofprefix.boot.tre. To obtain the bootstrap proportions they will then need to be read into PAUP* or a similar program to obtain a majority rule consensus. Note that it is probably safe to reduce the strictness of the termination conditions during bootstrapping (perhaps halve genthreshfortopoterm), which will greatly speed up the bootstrapping process with negligible effects on the results. When multiple bootstrapreps or multiple searchreps are specified we will automatically run garli multiple times in parallel. hidden_group4_bootstrap garli.conf perl !$user_conffile && $bootstrapreps > 1 && $searchreps_value == 1 4 perl "bootstrapreps = 1\\n" hidden_group5_bootstrap_multi garli.conf perl !$user_conffile && $bootstrapreps > 1 && $searchreps_value > 1 4 perl "bootstrapreps = $bootstrapreps\\n" outgroup_tax garli.conf 2 Outgroup taxa numbers, separated by spaces (outgroup) perl !$user_conffile perl "outgroup = $value\\n" The outgroup option allow for orienting tree topologies in a consistent way when they are written to file. Note that this has NO effect whatsoever on the actual inference and the specified outgroup is NOT constrained to be present in the inferred trees. If multiple outgroup taxa are specified and they do not form a monophyletic group, this setting will be ignored. If you specify a single outgroup taxon it will always be present, and the tree will always be consistently oriented. To specify an outgroup consisting of taxa 1, 3 and 5 the format is this: outgroup = 1 3 5 collapsebranches_g Collapse minimum length branches (effectively zero length) branch upon output of the final tree 0 perl !$user_conffile && $bootstrapreps == 0 collapsebranches garli.conf perl !$user_conffile && $collapsebranches_g 2 perl "collapsebranches = $collapsebranches_g\\n" ollapsebranches = (0 or 1, 1) Before version 1.0, all trees that are returned were fully resolved. This is true even if the maximum-likelihood estimate of some internal branch lengths are effectively zero (or GARLI's minimum, which is 1e-8). In such cases, collapsing the branch into a polytomy would be a better representation. Note that GARLI will never return a tree with an actual branch length of zero, but rather with its minimum value of 1.0e-8. The drawback of always returning fully resolved trees is that what is effectively a polytomy can be resolved in three ways, and different independent searches may randomly return one of those resolutions. Thus, if you compare the trees by topology only, they will look different. If you pay attention to the branch lengths and likelihood scores of the trees it will be apparent that they are effectively the same. I think that collapsing of branches is particularly important when bootstrapping, since no support should be given to a branch that doesn't really exist, i.e., that is a random resolution of a polytomy. Collapsing is also good when calculating tree to tree distances such as the symmetric tree distance, for example when calculating phylogenetic error to a known target tree. Zero-length branches would add to the distances (~error) although they really should not. constraintfile_control 2 Choose a Constraint File in Newick Format (constraintfile) constraint.txt To use this file correctly, you must also add the line constraintfile = constraint.txt to the [General] section of the garli.conf file, and remove any other lines containing constraintfile = perl $userconffilethere && defined $constraintfile_control The constraint file contains any topology constraint specifications, or none if there are no constraints. With Version 0.96, backbone constraints can be uploaded. A backbone constraint is the same as any other constraint file, but not all taxa need be represented. Format: Consider a dataset of 8 taxa, in which your constraint consists of grouping taxa 1, 3 and 5. You may specify either positive constraints (inferred tree MUST contain constrained group) or negative constraints (also called converse constraints, inferred tree CANNOT contain constrained group). These are specified with either a + or a - at the beginning of the constraint specification, for positive and negative constraints, respectively. For a positive constraint on a grouping of taxa 1, 3 and 5: +((1,3,5), 2, 4, 6, 7, 8); For a negative constraint on a grouping on taxa 1, 3 and 5: -((1,3,5), 2, 4, 6, 7, 8); (Note that there are many other equivalent parenthetical representations of this constraint.) GARLI also accepts another constraint format that may be easier to use in some cases. This involves specifying a single branch to be constrained with a string of * (asterisk) and . (period) characters, with one character per taxon. Each taxon specified with a * falls on one side of the constrained branch, and all those specified with a . fall on the other. This should be familiar to anyone who has looked at PAUP* bootstrap output. With this format, a positive constraint on a grouping of taxa 1, 3 and 5 would look like this: +*.*.*... or alternatively like this: +.*.*.*** With this format each line only designates a single branch, so multiple constrained branches may be specified as multiple lines in the file. constraintfile 2 garli.conf perl !$user_conffile perl (defined $constraintfile_control) ? "constraintfile = constraint.txt\\n" : "constraintfile = none\\n" model_initialization Initialization model_initialization_hidden garli.conf perl !$user_conffile 2 perl "refinestart = $refinestart\\n" . "randseed = $randseed\\n" refinestart Perform Initial Rough Optimization (refinestart) perl !$user_conffile 1 Specifies whether some initial rough optimization is performed on the starting branch lengths and alpha parameter. This is always recommended. randseed Random Seed ( -1 means it will be chosen for you) perl !$user_conffile -1 The random number seed used by the random number generator. Specify -1 to have a seed chosen for you. Specifying the same seed number in multiple runs will result in exactly identical runs, if all other parameters are also identical. model Model Parameters datatype_value The type of data (datatype) perl !$user_conffile nucleotide nucleotide aminoacid codon-aminoacid codon nucleotide "datatype = nucleotide\\n" aminoacid "datatype = aminoacid\\n" codon-aminoacid "datatype = codon-aminoacid\\n" codon "datatype = codon\\n" garli.conf 2 The codon-aminoacid datatype means that the data will be supplied as a nucleotide alignment, but will be internally translated and analyzed using an amino acid model. The codon and codon-aminoacid datatypes require nucleotide sequence that is aligned in the correct reading frame. In other words, all gaps in the alignment should be a multiple of 3 in length, and the alignment should start at the first position of a codon. If the alignment has extra columns at the start, middle or end, they should be removed or excluded with a Nexus exset (see the FAQ for an example of exset usage). The correct geneticcode must also be set. geneticcode_value Select the Genetic Code (geneticcode) perl !$user_conffile && ($datatype_value eq "codon" || $datatype_value eq "codon-aminoacid") perl "geneticcode = $value\\n" standard vertmito invertmito standard garli.conf 2 The codon-aminoacid datatype means that the data will be supplied as a nucleotide alignment, but will be internally translated and analyzed using an amino acid model. The codon and codon-aminoacid datatypes require nucleotide sequence that is aligned in the correct reading frame. In other words, all gaps in the alignment should be a multiple of 3 in length, and the alignment should start at the first position of a codon. If the alignment has extra columns at the start, middle or end, they should be removed or excluded with a Nexus exset (see the FAQ for an example of exset usage). The correct geneticcode must also be set. model_nucleotide Nucleotide Model JC (Jukes-Cantor model): Rate Matrix = 1 rate, Base Frequencies = equal K2P (Kimura 2-Parameter model): Rate Matrix = 2 rate, Base Frequencies = equal F81 (Felsenstein 1981 model): Rate Matrix = 1 rate, Base Frequencies = estimate HKY (Hasegawa, Kishino and Yano model): Rate Matrix = 2 rate, Base Frequencies = estimate GTR (General Time-Reversible model): Rate Matrix = 6 rate, Base Frequencies = estimate d_ratematrix garli.conf Rate Matrix (ratematrix) perl !$user_conffile && $datatype_value eq "nucleotide" 1rate 2rate 6rate fixed custom_string 1rate "ratematrix = $value\\n" 2rate "ratematrix = $value\\n" 6rate "ratematrix = $value\\n" fixed "ratematrix = $value\\n" custom_string "ratematrix = ($ACsubrates $AGsubrates $ATsubrates $CGsubrates $CTsubrates $GTsubrates)\\n" 6rate garli.conf 2 Parameters for any submodel of the GTR model may be estimated. The format for specifying this is very similar to that used in the rclass setting of PAUP*. Within parentheses, six letters are specified, with spaces between them. The six letters represent the rates of substitution between the six pairs of nucleotides, with the order being A-C, A-G, A-T, C-G, C-T and G-T. Letters within the parentheses that are the same mean that a single parameter is shared by multiple nucleotide pairs. For example, ratematrix = (a b a a b a) would specify the HKY 2-rate model (equivalent to ratematrix = 2rate). This entry, ratematrix = (a b c c b a) would specify 3 estimated rates of substitution, with one rate shared by A-C and G-T substitutions, another rate shared by A-G and C-T substitutions, and the final rate shared by A-T and C-G substitutions. ACsubrates User specified AC substitution rates (custom ratematrix) perl $datatype_value eq "nucleotide" && $d_ratematrix eq "custom_string" perl "" AGsubrates User specified AG substitution rates (custom ratematrix) perl $datatype_value eq "nucleotide" && $d_ratematrix eq "custom_string" perl "" ATsubrates User specified AT substitution rates (custom ratematrix) perl $datatype_value eq "nucleotide" && $d_ratematrix eq "custom_string" perl "" CGsubrates User specified CG substitution rates (custom ratematrix) perl $datatype_value eq "nucleotide" && $d_ratematrix eq "custom_string" perl "" CTsubrates User specified CT subsitution rates (custom ratematrix) perl $datatype_value eq "nucleotide" && $d_ratematrix eq "custom_string" perl "" GTsubrates User specified GT substitution rates (custom ratematrix) perl $datatype_value eq "nucleotide" && $d_ratematrix eq "custom_string" perl "" d_statefrequencies garli.conf Base Frequencies (statefrequencies) perl !$user_conffile && $datatype_value eq "nucleotide" perl "statefrequencies = $value\\n" equal empirical estimate fixed estimate 2 d_invariantsites garli.conf Proportion of invariant sites (invariantsites) perl !$user_conffile && $datatype_value eq "nucleotide" perl "invariantsites = $value\\n" none estimate estimate 2 Specifies whether a parameter representing the proportion of sites that are unable to change (i.e. have a substitution rate of zero) will be included. This is typically referred to as invariant sites, but would better be termed invariable sites. d_ratehetmodel garli.conf The model of rate heterogeneity (ratehetmodel) perl !$user_conffile && $datatype_value eq "nucleotide" perl ($d_ratehetmodel eq "none") ? "numratecats = 1\\nratehetmodel = $value\\n" : "numratecats = $d_numratecats\\nratehetmodel = $value\\n" none gamma gamma 2 The model of rate heterogeneity assumed. gammafixed requires that the alpha shape parameter is provided, and a setting of gamma estimates it. d_numratecats Number of rate categories (numratecats) perl !$user_conffile && $datatype_value eq "nucleotide" && $d_ratehetmodel eq "gamma" 4 :Number of rate categories: must be an integer between 1 and 20 perl $d_numratecats < 2 || $d_numratecats > 20 garli.conf 2 The number of categories of variable rates (not including the invariant site class if it is being used). Must be set to 1 if ratehetmodel is set to none. Note that runtimes and memory usage scale linearly with this setting. model_protein Protein Model dayhoff Dayhoff, Schwartz and Orcutt. 1978 jones Jones, Taylor and Thornton (JTT), 1992 WAG Whelan and Goldman, 2001 mtREV Adachi and Hasegawa, 1996 mtmam Yang, Nielsen and Hasegawa. 1998 p_ratematrix 2 Protein Rate Matrix (ratematrix) perl !$user_conffile && ($datatype_value eq "aminoacid" || $datatype_value eq "codon-aminoacid") perl "ratematrix = $value\\n" wag poisson jones dayhoff wag mtmam mtrev garli.conf You should use the matrix that gives the best likelihood, and could use a program like PROTTEST (very much like MODELTEST, but for amino acid models) to determine which fits best for your data. Poisson assumes a single rate of substitution between all amino acid pairs, and is a very poor model. p_statefrequencies garli.conf 2 Amino Acid Frequencies (statefequencies) perl !$user_conffile && ($datatype_value eq "aminoacid" || $datatype_value eq "codon-aminoacid") perl "statefrequencies = $value\\n" equal empirical estimate fixed jones dayhoff wag mtmam mtrev empirical Specifies how the equilibrium state frequencies of the 20 amino acids are treated. The empirical option fixes the frequencies at their observed proportions (when describing a model this is often termed +F). p_invariantsites garli.conf 2 Proportion of invariable sites (invariantsites) perl !$user_conffile && ($datatype_value eq "aminoacid" || $datatype_value eq "codon-aminoacid") perl "invariantsites = $value\\n" none estimate estimate Specifies whether a parameter representing the proportion of sites that are unable to change (i.e. have a substitution rate of zero) will be included. This is typically referred to as invariant sites, but would better be termed invariable sites. p_ratehetmodel 2 Model of rate heterogeneity (ratehetmodel) perl !$user_conffile && ($datatype_value eq "aminoacid" || $datatype_value eq "codon-aminoacid") perl ($p_ratehetmodel eq "none") ? "numratecats = 1\\nratehetmodel = $value\\n" : "numratecats = $p_numratecats\\nratehetmodel = $value\\n" none gamma gamma garli.conf The model of rate heterogeneity assumed. gammafixed requires that the alpha shape parameter is provided, and a setting of gamma estimates it. p_numratecats garli.conf 2 Number of rate categories (numratecats; set at no more than 8) perl !$user_conffile && $p_ratehetmodel eq "gamma" && ($datatype_value eq "aminoacid" || $datatype_value eq "codon-aminoacid") 4 Number of rate categories: must be an integer between 1 and 20 perl $p_numratecats < 2 || $p_numratecats > 20 The number of categories of variable rates (not including the invariant site class if it is being used). Must be set to 1 if ratehetmodel is set to none. Note that runtimes and memory usage scale linearly with this setting. model_codon Codon Model The codon models are built with three components: (1) parameters describing the process of individual nucleotide substitutions, (2) equilibrium codon frequencies, and (3) parameters describing the relative rate of nonsynonymous to synonymous substitutions. The nucleotide substitution parameters within the codon models are exactly the same as those possible with standard nucleotide models in GARLI, and are specified with the ratematrix configuration entry. Thus, they can be of the 2rate variety (inferring different rates for transitions and transversions, K2P or HKY-like), the 6rate variety (inferring different rates for all nucleotide pairs, GTR-like) or any other sub-model of GTR. The options for codon frequencies are specified with the statefrequencies configuration entry. The options are to use equal frequencies (not a good option), the frequencies observed in your dataset (termed empirical in GARLI), or the codon frequencies implied by the F1x4 or F3x4 methods (using PAML's terminology). These last two options calculate the codon frequencies as the product of the frequencies of the three nucleotides that make up each codon. In the F1x4 case the nucleotide frequencies are those observed in the dataset across all codon positions, while the F3x4 option uses the nucleotide frequencies observed in the data at each codon position separately. The final component of the codon models is the nonsynonymous to synonymous relative rate parameters (aka dN/dS or omega parameters). The default is to infer a single dN/dS value. Alternatively, a model can be specified that infers a given number of dN/dS categories, with the dN/dS values and proportions falling in each category estimated (ratehetmodel = nonsynonymous). This is the discrete or M3 model in PAML's terminology. From the Author: No stop codons under the chosen genetic code are allowed. These are: standard code (TAG, TAA and TGA); vertebrate mitochondria (TAG, TAA, AGA and AGG); invertebrate mitochondria (TAG and TAA). One might argue that stop codons should just be ignored and treated as missing, but this can be dangerous. Sometimes they will come from a sequencing error, but more often from an alignment problem, an incorrectly chosen genetic code, or a sequence that is not really coding (e.g. an intron). In any case, error should be examined and resolved consciously by the user. One thing to note is that codon models for tree inference require that you align the protein coding sequences along a correct reading frame; (e.g., gaps of 1 or 2 bases will impair the analysis). Maintaining the reading frame makes alignment much easier even if your analysis will be at the nucleotide level. Just running sequences through a sequence alignment program without looking is almost guaranteed to return an alignment that will not work for codon based inference. The other important restriction to note is that GARLI expects the alignment to begin on the first base of a codon. It can't figure out where the reading frame is, so if the alignment starts with a partial codon it needs to be removed or excluded from the alignment. Version 0.96 does allow normal NEXUS exclusions through an assumptions block, so the following would work to exclude the first two bases of an alignment and tell GARLI that the reading frame starts on the third. begin assumptions; exset * myexset = 1 2; end;. c_ratematrix garli.conf 2 Codon Rate Matrix (ratematrix) perl !$user_conffile && $datatype_value eq "codon" perl "ratematrix = $value\\n" 2rate 1rate 2rate 6rate fixed custom string This determines the relative rates of nucleotide substitution assumed by the codon model. The options are exactly the same as those allowed under a normal nucleotide model. A codon model with ratematrix = 2rate specifies the standard Goldman and Yang (1994) model, with different substitution rates for transitions and transversions. c_statefrequencies garli.conf 2 Codon Frequencies (statefequencies) perl !$user_conffile && $datatype_value eq "codon" perl "statefrequencies = $value\\n" equal empirical f1x4 f3x4 f3x4 The options are to use equal codon frequencies (not a good option), the frequencies observed in your dataset (termed empirical in GARLI), or the codon frequencies implied by the F1x4 or F3x4 methods (using PAML's terminology). These last two options calculate the codon frequencies as the product of the frequencies of the three nucleotides that make up each codon. In the F1x4 case the nucleotide frequencies are those observed in the dataset across all codon positions, while the F3x4 option uses the nucleotide frequencies observed in the data at each codon position separately. c_ratehetmodel garli.conf 2 dN/dS categories (or Omega) (ratehetmodel) perl !$user_conffile && $datatype_value eq "codon" none nonsynonymous none "ratehetmodel = none\\ninvariantsites = none\\nnumratecats = 1\\n" nonsynonymous "ratehetmodel = nonsynonymous\\ninvariantsites = none\\n" none For codon models, the default is to infer a single dN/dS parameter. Alternatively, a model can be specified that infers a given number of dN/dS categories, with the dN/dS values and proportions falling in each category estimated (ratehetmodel = nonsynonymous). This is the discrete or M3 model of Yang et al. (2000). c_numratecats Number of dN/dS parameter categories (numratecats) garli.conf perl !$user_conffile && $datatype_value eq "codon" && $c_ratehetmodel eq "nonsynonymous" perl "numratecats = $value\\n" 3 Number of rate categories: must be an integer between 1 and 8 perl $c_numratecats < 1 || $c_numratecats > 8 2 When ratehetmodel = nonsynonymous, this is the number of dN/dS parameter categories. algorithm_population Population poplulation_hidden garli.conf perl !$user_conffile 4 perl "selectionintensity = $selectionintensity\\n" . "nindivs = $nindivs\\n" . "holdover = $holdover\\n" selectionintensity Selection Intensity (0.01 to 5.0) (selectionintensity) perl !$user_conffile 0.5 Selection Intensity must be between 0.01 - 5.0 perl $selectionintensity < 0.01 || $selectionintensity > 5.0 Controls the strength of selection, with larger numbers denoting stronger selection. The relative probability of reproduction of two individuals depends on the difference in their log likelihoods (delta lnL) and is formulated very similarly to the procedure of calculating Akaike weights. The relative probability of reproduction of the less fit individual is equal to: In general this setting does not seem to have much of an effect on the progress of a run. In theory higher values should cause scores to increase more quickly, but make the search more likely to be entrapped in a local optimum. The following table gives the relative probabilities of reproduction for different values of the selection intensity when the difference in log likelihood is 1.0 Selection intensity Ratio of probabilities of reproduction 0.05 0.95:1.0 0.1 0.90:1.0 0.25 0.78:1.0 0.5 0.61:1.0 0.75 0.47:1.0 1 0.37:1.0 2 0.14:1.0 nindivs Number of Individuals (nindivs 2 to 100) perl !$user_conffile 4 Number of individuals must be between 2 - 100 perl $nindivs < 2 || $nindivs > 100 The number of individuals in the population. This may be increased, but generally seems to slow the rate of score increase. holdover perl !$user_conffile 1 algorithm_brlen Branch-length Optimization brlen_hidden garli.conf perl !$user_conffile 4 perl "startoptprec = $startoptprec\\n" . "minoptprec = $minoptprec\\n" . "numberofprecreductions = $numberofprecreductions\\n" startoptprec Starting Precision (startoptprec: 0.005 - 5.0) 0.5 perl !$user_conffile Starting Precision must be between 0.005 - 5.0 perl $startoptprec < 0.005 || $startoptprec > 5.0 minoptprec Minimum Precision (minoptprec: 0.001 - 0.01) perl !$user_conffile 0.01 Minimum Precision must be between 0.001 - 0.01 perl $minoptprec < 0.001 || $minoptprec > 0.01 numberofprecreductions Number of Precision Reductions (0 - 100) perl !$user_conffile 20 Number of Precision Reductions must be between 0 - 100 perl $minoptprec < 0 || $minoptprec > 100 Specify the number of steps that it will take for the optimization precision to decrease from startoptprec to minoptprec. In version 0.95, the reduction from startoptprec to minoptprec is linear, rather than geometric. algorithm_mutation_prior_weighting Mutation Prior Weighting prior_weighting_hidden garli.conf perl !$user_conffile 4 perl "modweight = $modweight\\n" . "brlenweight = $brlenweight\\n" . "topoweight = $topoweight\\n" . "randnniweight = $randnniweight\\n" . "randsprweight = $randsprweight\\n" . "limsprweight = $limsprweight\\n" modweight Model Mutations (modweight) perl !$user_conffile 0.05 Model Mutations must be 0 or greater. perl $modweight < 0 The prior weight assigned to the class of model mutations. Note that setting this at 0.0 fixes the model during the run. brlenweight Branch-length Mutations (brlenweight) perl !$user_conffile 0.2 Branch-length Mutations must be 0 or greater. perl $brlenweight < 0 The prior weight assigned to branch-length mutations. topoweight garli.conf perl !$user_conffile 4 All Topology Mutations (topoweight) 1.0 :All Topology Mutations: must be 0 or greater. perl $topoweight < 0 The prior weight assigned to the class of topology mutations (NNI, SPR and limSPR). randnniweight NNI Mutations (randnniweight) perl !$user_conffile 0.1 NNI Mutations must be 0 or greater. perl $randnniweight < 0 The prior weight assigned to NNI mutations. randsprweight SPR Mutations (randsprweight) perl !$user_conffile 0.3 SPR Mutations must be 0 or greater. perl $randsprweight < 0 randsprweight (0 to infinity, 0.3) -The prior weight assigned to random SPR mutations. For very large datasets it is often best to set this to 0.0, as random SPR mutations essentially never result in score increases. limsprweight Limited SPR Mutations (limsprweight) perl !$user_conffile 0.6 Limited SPR Mutations must be 0 or greater. perl $limsprweight < 0 The prior weight assigned to SPR mutations with the reconnection branch limited to being a maximum of limsprrange branches away from where the branch was detached. algorithm_mutation_details Set Mutation Details mutation_details_hidden garli.conf perl !$user_conffile 4 perl "limsprrange= $limsprrange\\n" . "uniqueswapbias = $uniqueswapbias\\n" . "distanceswapbias = $distanceswapbias\\n" limsprrange Max Limited SPR Branch Movement (limsprrange) perl !$user_conffile 6 Limited SPR Branch Movement must be 0 or greater. perl $limsprrange < 0 The maximum number of branches away from its original location that a branch may be reattached during a limited SPR move. Setting this too high (> 10) can seriously degrade performance. uniqueswapbias Unique Swap Bias (uniqueswapbias) perl !$user_conffile 0.1 Unique Swap Bias must be between 0.01 - 1.0 perl $uniqueswapbias < 0.01 || $uniqueswapbias > 1.0 distanceswapbias Distance Swap Bias (distanceswapbias) perl !$user_conffile 1.0 general_logs Configure Logs saveevery_g Save best tree with interval (saveevery) perl !$user_conffile 100 Save best tree with interval must be greater than 1 perl $value < 1 saveevery garli.conf perl !$user_conffile 2 perl "saveevery = $saveevery_g\\n" outputcurrentbesttopology Output current best tree at "savevery" interval (outputcurrentbesttopology) garli.conf perl !$user_conffile 2 perl ($value) ? "outputcurrentbesttopology = 1\\n" : "outputcurrentbesttopology = 0\\n" 1 logevery_g Save best score with interval (logevery) perl !$user_conffile 10 Save best score with interval must be greater than 1 perl $value < 1 The frequency with which the best score is written to the log file. logevery garli.conf perl !$user_conffile 2 perl "logevery = $logevery_g\\n" outputeachbettertopology_g Save each improved topology (outputeachbettertopology; can result in a very large file) perl !$user_conffile 0 If true, each new topology encountered with a better score than the previous best is written to file. In some cases this can result in really big files, possibly hundreds of MB, especially for random starting topologies on large datasets. Note that this file is interesting to get an idea of how the topology changed as the searches progressed, but the collection of trees should NOT be interpreted in any meaningful way. This option is not available while bootstrapping. outputeachbettertopology garli.conf perl !$user_conffile 2 perl "outputeachbettertopology = $outputeachbettertopology_g\\n" inferinternalstateprobs_g Output state probabilities of internal nodes (inferinternalstateprobs) perl !$user_conffile || $bootstrapreps == 0 0 inferinternalstateprobs garli.conf 4 perl !$user_conffile perl "inferinternalstateprobs = $inferinternalstateprobs_g\\n" Use this flag to have GARLI infer the marginal posterior probability of each character at each internal node. This is done at the very end of the run, just before termination. The results are output to a file named ofprefix.internalstates.log. outputphyliptree_g Output PHYLIP-format tree (outputphyliptree) 0 perl !$user_conffile outputphyliptree garli.conf perl !$user_conffile 2 perl "outputphyliptree = $outputphyliptree_g\\n" outputmostlyuselessfiles_g Output fate, problog, and swaplog files perl !$user_conffile 0 outputmostlyuselessfiles garli.conf perl !$user_conffile 2 perl "outputmostlyuselessfiles = $outputmostlyuselessfiles_g\\n" Whether to output three files of little general interest: the fate, problog, and swaplog files. The fate file shows the parentage, mutation types and scores of every individual in the population during the entire search. The problog shows how the proportions of the different mutation types changed over the course of the run. The swaplog shows the number of unique swaps and the number of total swaps on the current best tree over the course of the run. general_run Configure Run Termination hidden_group2 garli.conf perl !$user_conffile 2 perl "enforcetermconditions = $enforcetermconditions\\n" . "genthreshfortopoterm = $genthreshfortopoterm\\n" . "significanttopochange = $significanttopochange\\n" . "scorethreshforterm = $scorethreshforterm\\n" hidden_group4 garli.conf perl !$user_conffile 4 perl "stopgen = $stopgen\\n" . "stoptime = $stoptime\\n" enforcetermconditions Automatically terminate run (enforcetermconditions; recommended) 1 perl !$user_conffile genthreshfortopoterm perl !$user_conffile && $enforcetermconditions Generations without improving topology (genthreshfortopoterm) 20000 Generations without improving topology must be a positive integer perl $value < 0 This specifies the first part of the termination condition. When no new significantly better scoring topology see significanttopochange below) has been encountered in greater than this number of generations, this condition is met. Increasing this parameter may improve the lnL scores obtained (especially on large datasets), but will also increase runtimes. significanttopochange perl !$user_conffile && $enforcetermconditions lnL increase for significantly better topology (significanttopochange) 0.01 lnL increase for significantly better topology must be a positive number perl $value < 0 The lnL increase required for a new topology to be considered significant as far as the termination condition is concerned. It probably doesn't need to be played with, but you might try increasing it slightly if your runs reach a stable score and then take a very long time to terminate due to very minor changes in topology. scorethreshforterm perl !$user_conffile && $enforcetermconditions Score improvement threshold (scorethreshforterm) 0.05 Score improvement threshold must be a positive number. perl $value < 0 The second part of the termination condition. When the total improvement in score over the last intervallength x intervalstostore generations (default is 500 generations, see below) is less than this value, this condition is met. This does not usually need to be changed. stopgen Limit generations to maximum of (stopgen); default =214783646 perl !$user_conffile 214783646 Limit generations to maximum of must be a positive number; unless you are experienced, we recommend the default, 214783646 perl $value < 1 The maximum number of generations to run. Note that this supersedes the automated stopping criterion (see enforcetermconditions above), and should therefore be set to a very large value if automatic termination is desired. stoptime Limit run time to maximum of (stoptime, in seconds) default=214783646 214783646 perl !$user_conffile :Limit run time to maximum of: must be a positive number; unless you are experienced, we recommend the default, 214783646 perl $value < 1 The maximum number of seconds for the run to continue. Note that this supersedes the automated stopping criterion (see enforcetermconditions above), and should therefore be set to a very large value if automatic termination is desired. resampleproportion Use Resampling where the pseudoreplicate datasets have a different number of alignment columns than the real data (resampleproportion) perl !$user_conffile && $bootstrapreps > 0 resampleproportion_val Enter the value for resampling proportion (resampleproportion) perl !$user_conffile && $resampleproportion 4 1.0 perl "resampleproportion = $resampleproportion_val\\n" Resample proportion must be a positive number less than 10 perl $value < 0 || $value > 10 When bootstrapreps is greater than 0, this setting allows for bootstrap-like resampling, but with the pseudoreplicate datasets having a different number of alignment columns than the real data. Setting values less than 1.0 is akin to jackknifing writecheckpoints garli.conf perl !$user_conffile 2 Write Checkpoints (writecheckpoints) perl "writecheckpoints = $vdef\\n" 0 restart garli.conf 2 perl !$user_conffile Restart (restart) perl "restart = $vdef\\n" 0 holdoverpenalty garli.conf 4 Holdover Penalty (holdoverpenalty) perl !$user_conffile perl "holdoverpenalty = $vdef\\n" 0 treerejectionthreshold garli.conf 4 treerejectionthreshold perl !$user_conffile perl "treerejectionthreshold = $vdef\\n" 50.0 intervallength garli.conf perl !$user_conffile 4 intervallength perl "intervallength = $vdef\\n" 100 intervalstostore garli.conf perl !$user_conffile 4 intervalstostore perl "intervalstostore = $vdef\\n" 5 meanbrlenmuts garli.conf perl !$user_conffile 4 Mean Branch-length mutations (1 - # taxa) perl "meanbrlenmuts = $vdef\\n" 5 gammashapebrlen garli.conf perl !$user_conffile 4 gammashapebrlen perl "gammashapebrlen = $vdef\\n" 1000 gammashapemodel garli.conf perl !$user_conffile 4 gammashapemodel perl "gammashapemodel = $vdef\\n" 1000