jModelTest2 on XSEDE 2.1.6 Statistical selection of best-fit models of nucleotide substitution, run on XSEDE Diego Darriba and David Posada Darriba D, Taboada GL, Doallo R, and Posada D. (2012) jModelTest 2: more models, new heuristics and parallel computing. Nature Methods 9(8), 772. Phylogeny / Alignment jmodeltest2_expanse_xsede jmodeltest2_invoke perl ($set_subschemes*$uneq_basefmodels*$invar_models*$include_ratevar) <= 24 perl "" 0 jmodeltest2_invoke2 perl ($set_subschemes*$uneq_basefmodels*$invar_models*$include_ratevar) > 24 && $set_subschemes*$uneq_basefmodels*$invar_models*$include_ratevar < 56 perl "" 0 jmodeltest2_invoke3 perl ($set_subschemes*$uneq_basefmodels*$invar_models*$include_ratevar) > 55 perl "" 0 number_nodes 2 perl $set_subschemes*$uneq_basefmodels*$invar_models*$include_ratevar <= 24 scheduler.conf perl "nodes=1\\n" . "node_exclusive=0\\n" . "mem=30G\\n" . "cpus-per-task=16\\n" . "threads_per_process=16\\n" number_nodes2 2 perl $set_subschemes*$uneq_basefmodels*$invar_models*$include_ratevar > 24 && $set_subschemes*$uneq_basefmodels*$invar_models*$include_ratevar < 56 scheduler.conf perl "nodes=1\\n" . "node_exclusive=0\\n" . "mem=46G\\n" . "cpus-per-task=24\\n" . "threads_per_process=24\\n" number_nodes3 2 perl $set_subschemes*$uneq_basefmodels*$invar_models*$include_ratevar > 55 scheduler.conf perl "nodes=1\\n" . "node_exclusive=0\\n" . "mem=61G\\n" . "cpus-per-task=32\\n" . "threads_per_process=32\\n" infile Input Alignment infile.phy perl "-d infile.phy" 1 runtime 1 scheduler.conf Maximum Hours to Run (up to 168 hours) 0.5 The maximum hours to run must be less than 168 perl $runtime > 168.0 The maximum hours to run must be greater than 0.05 perl $runtime < 0.05 perl "runhours=$value\\n" The job will run on 16 processors as configured. If it runs for the entire configured time, it will consume 16 x $runtime cpu hours perl $set_subschemes*$uneq_basefmodels*$invar_models*$include_ratevar <= 24 The job will run on 24 processors as configured. If it runs for the entire configured time, it will consume 24 x $runtime cpu hours perl $set_subschemes*$uneq_basefmodels*$invar_models*$include_ratevar > 24 && $set_subschemes*$uneq_basefmodels*$invar_models*$include_ratevar < 56 The job will run on 32 processors as configured. If it runs for the entire configured time, it will consume 32 x $runtime cpu hours perl $set_subschemes*$uneq_basefmodels*$invar_models*$include_ratevar > 55 Estimate the maximum time your job will need to run. We recommend testing 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. name_output Specify the name of your output file (-o) perl !$convert_phylip perl (defined $value) ? "-o $value":"" output.txt 3 name_suffix Specify a suffix for your log file (-n) perl !$convert_phylip perl (defined $value) ? "-n $value":"" 4 convert_phylip Convert input to Phylip format (only) perl ($value) ? "-getPhylip":"" 0 2 choose_models Configure Model Search perl !$convert_phylip information_criterion Clustering search (-H) perl $set_subschemes eq "203" AIC AICc BIC perl "-H $value" BIC 25 This is used only with 203 searches. AIC=Akaike Information Criterion; AICc = corrected Akaike Information Criterion; BIC = Bayesian Information Criterion threshold_heuristic Perform threshold heuristic search. (-G; default = 0.1) perl !$convert_phylip perl (defined $value) ? "-G $value":"" 13 set_subschemes Set the number of substitution schemes (-s) perl !$convert_phylip 3 5 7 11 203 perl " -s $value" 11 5 perl !defined $set_subschemes This flag sets the number of substitution schemes. 3 = JC/F81 K80/HKY SYM/GTR; 5 = JC/F81, K80/HKY, TrNef/TrN, TPM1/TPM1uf, SYM/GTR 7 = JC/F81, K80/HKY, TrNef/TrN, TPM1/TPM1uf, TIM1ef/TIM1, TVMef/TVM, SYM/GTR; 11 = all named models; 203 = All possible GTR submatrices uneq_basefmodels Include models with unequal base frequencies. (-f) perl !$convert_phylip 1 2 1 "" 2 "-f" 2 7 Please specify whether or not to use the -f option perl !defined $uneq_basefmodels invar_models Include models with a proportion invariable sites. (-i) perl !$convert_phylip 1 2 1 "" 2 "-i" 2 Please specify whether or not to include models with invariable sites (-i) perl !defined $invar_models 8 include_ratevar Include models with rate variation among sites? 1 2 2 Please specify whether or not to include models with rate variation perl !defined $include_ratevar numratecat_models Include models with rate variation among sites, set number of categories (-g; default = 4) perl !$convert_phylip perl (defined $value) ? "-g $value":"" 4 9 confidenceint_models Confidence interval for the model selection process (-c; default = 100) perl !$convert_phylip perl (defined $value) ? "-c $value":"" 18 configure_treesearch Define Tree Search set_basetree Base tree topology for likelihood calculations. (-t) perl !$convert_phylip fixed BIONJ ML perl " -t $value" ML This option lets you choose the base tree topology for liklihood calculations. The choices are: Fixed BIONJ: Fixed BIONJ topology from Jukes-Cantor model; Neighor Joining: Use Neighbor-Joining topology for each model; and Maximum Likelihood: use the maximum liklihood topology for each model 10 def_topsearch Set tree topology search operation option for Maximum-Likelihood search (-S) perl !$convert_phylip && $set_basetree eq "ML" && !defined $user_fixedtree NNI SPR BEST BEST perl " -S $value" 10 by default, jModelTest uses Maximum-Likelihood topologies as the base trees for the model optimization, and checks both Nearest Neighbour Interchange (fast)and Subtree Pruning and Regrafting (slower) algorithms for the topology search. This obtains the most accurate results, but it is also the most time consuming operation. According to the size of the input alignment, one can directly select one of the algorithms saving time in the computations. As a general rule, for a small number of taxa NNI algorithm would work better, as well as SPR is more suitable for a large number of taxa. The tree search operation can be set with “-S” argument (e.g., -t ML -S NNI). user_fixedtree Select a User-defined fixed tree for likelihood calculations (-u) perl !$convert_phylip userfixedtree.tre perl "-u userfixedtree.tre" 8 selection_criteria Information Criteria criteria_1 Select Information Criterion (choose all needed) perl !$convert_phylip -AIC -AICc -BIC -DT perl (defined $value) ? "$value":"" -AIC -AICc -BIC -DT 10 Please select at least one Information Criterion (all four is typical) perl !defined $criteria_1 ' ' confidence_selection Confidence level for the hLRTs (-h; default is 0.01) perl !$convert_phylip perl (defined $value) ? "-h $value":"" 13 parameter_importances Calculate the parameter importances (-p) perl !$convert_phylip perl ($value) ? "-p":"" 1 16 modeland_importances Do model averaging and parameter importances (-v) perl !$convert_phylip perl ($value) ? "-v":"" 0 17 print_paup Print out the PAUP block (-w) perl !$convert_phylip perl ($value) ? "-w":"" 1 You can append the PAUP block produced here if you want to load the selected model and associated estimates in PAUP> 20 estimate_modelavg Estimate model-averaged phylogeny for each active criterion (-a) perl !$convert_phylip perl ($value) ? "-a":"" 1 19 strict_consensus Use strict consensus type for model-averaged phylogeny (-z) perl !$convert_phylip perl ($value) ? "-z":"" 0 14 Default for this parameter is Majority Rule calculate_delta Calculate delta AIC,AICc,BIC against unconstrained likelihood. (-uLnL) perl !$convert_phylip perl ($value) ? "-uLnL":"" 0 19 hRLT_params hRLT parameters hierarchical_likelihood Perform hierarchical likelihood ratio tests (-hLRT) perl !$convert_phylip && $set_subschemes ne "203" &&($set_basetree eq "fixed" || defined $user_fixedtree) perl ($value) ? "-hLRT":"" 0 13 The -hrlt option not available for the 203 subsitution scheme. It requires an uploaded user-defined topology, or a fixed BIONJ-JC tree. specify_hyporder Specify the hypothesis order (-O; default ftvwxgp) perl $hierarchical_likelihood perl (defined $value) ? "-O $value":"" ftvwxgp 14 This option allows you to set the hypothesis order as a string of small case letters. The options are as follows: f:frequencies; t:transition/transversion ratio; v: 2ti4tv for subst=3 or 2ti for subst¿3; w: 2tv; x: 4tv; g: gamma; p: proportion of invariable sites. backward_selection Backward selection for the hLRT (-r) perl $hierarchical_likelihood perl ($value) ? "-r":"" 0 13 dynamical_likelhood Perform dynamical likelihood ratio tests. (-dLRT) perl $hierarchical_likelihood perl ($value) ? "-dLRT":"" 0 12 all_results *