Description
Infers a species network(s) using maximum likelihood, where K-fold cross-validation is used to account for model complexity. Only topologies of gene trees are used in the inference.
Usage
InferNetwork_ML_CV (gene_tree_ident1 [, gene_tree_ident2...]) numReticulations [-cv numFolds] [-a taxa map] [-b threshold] [-s startingNetwork] [-h {s1 [,s2...]}] [-w (w1,w2,w3,w4)] [-f maxFailure] [-x numRuns] [-m maxNetExamined] [-d maxDiameter] [-p (rel,abs)] [-r maxRounds] [-t maxTryPerBr] [-i improveThreshold] [-l maxBL] [-pl numProcessors] [-di] [result output file]
gene_tree_ident1 [, gene_tree_ident2...] | Comma delimited list of gene tree identifiers. | mandatory |
numReticulations | Maximum number of reticulations to added. | mandatory |
-cv numFolds | Number of folds in K-fold cross-validation. Default value is 10. | optional |
-b threshold | Gene trees bootstrap threshold. | optional |
-a taxa map | Gene tree / species tree taxa association. | optional |
-s startingNetwork | Specify the network to start search. Default value is the optimal MDC tree. | optional |
-h {s1 [, s2...]} | A set of specified hybrid species. The size of this set equals the number of reticulation nodes in the inferred network. | optional |
-w (w1, w2, w3, w4) | The weights of operations for network arrangement during the network search. Default value is (0.15, 0.15, 0.2, 0.5). | optional |
-f maxFailure | The maximum number of consecutive failures before the search terminates. Default value is 100. | optional |
-x numRuns | The number of runs of the search. Default value is 5. | optional |
-m maxNetExamined | Maximum number of network topologies to examined. Default value is infinity. | optional |
-d maxDiameter | Maximum diameter to make an arrangement during network search. Default value is infinity. | optional |
-p (rel, abs) | The original stopping criterion of Brent’s algorithm. Default value is (0.01, 0.001). | optional |
-r maxRound | Maximum number of rounds to optimize branch lengths for a network topology. Default value is 100. | optional |
-t maxTryPerBr | Maximum number of trial per branch in one round to optimize branch lengths for a network topology. Default value is 100. | optional |
-i improveThreshold | Minimum threshold of improvement to continue the next round of optimization of branch lengths. Default value is 0.001. | optional |
-l maxBL | Maximum branch lengths considered. Default value is 6. | optional |
-pl numProcessors | Number of processors if you want the computation to be done in parallel. Default value is 1. | optional |
-di | Output the Rich Newick string of the inferred network that can be read by Dendroscope . | optional |
result output file | Optional file destination for command output. | optional |
By default, 10-fold cross-validation is used to account for model complexity. But users can change the number of folds using option -cv.
See command InferNetwork_ML for all other parameters.
Examples
#NEXUS BEGIN TREES; TREE gt0 = (D:7.07072,((C:3.56753,B:3.56753):1.76822,A:5.33575):1.73497); TREE gt1 = (((B:1.97661,C:1.97661):2.23864,A:4.21524):1.77987,D:5.99511); TREE gt2 = ((C:4.31675,(B:3.14621,A:3.14621):1.17054):2.09695,D:6.4137); TREE gt3 = ((D:5.83927,A:5.83927):0.566624,(B:1.80987,C:1.80987):4.59603); TREE gt4 = ((D:5.77537,(B:1.77451,C:1.77451):4.00086):0.810136,A:6.58551); TREE gt5 = (D:6.80413,(A:3.82444,(C:2.31671,B:2.31671):1.50773):2.97969); TREE gt6 = (D:7.61541,(C:4.41986,(A:2.52336,B:2.52336):1.8965):3.19554); TREE gt7 = ((A:4.99068,(C:3.03372,B:3.03372):1.95696):0.782212,D:5.77289); TREE gt8 = (D:5.95232,((C:1.86462,B:1.86462):3.20304,A:5.06765):0.88467); TREE gt9 = ((A:3.84332,(C:3.74007,B:3.74007):0.103251):4.82743,D:8.67075); TREE gt10 = ((A:4.19291,(C:1.73235,B:1.73235):2.46056):2.60752,D:6.80043); TREE gt11 = (A:7.35563,((B:1.7592,C:1.7592):4.0683,D:5.8275):1.52813); TREE gt12 = (((B:3.14432,A:3.14432):0.613233,C:3.75755):1.95882,D:5.71638); TREE gt13 = (((C:2.6677,B:2.6677):2.66994,A:5.33764):4.71766,D:10.0553); TREE gt14 = ((C:4.00286,(B:3.0957,A:3.0957):0.907165):1.94607,D:5.94893); TREE gt15 = (D:5.89168,(A:3.77037,(B:2.13754,C:2.13754):1.63283):2.12131); TREE gt16 = (D:6.43412,(A:3.85214,(B:2.6542,C:2.6542):1.19794):2.58198); TREE gt17 = (D:6.27103,((B:1.80021,C:1.80021):1.91286,A:3.71307):2.55796); TREE gt18 = (((C:1.72131,B:1.72131):4.04022,A:5.76154):2.49476,D:8.25629); TREE gt19 = (((B:1.8262,C:1.8262):2.49809,A:4.32429):3.80882,D:8.13311); TREE gt20 = (D:6.95436,((B:2.60072,C:2.60072):1.85771,A:4.45842):2.49594); END; BEGIN PHYLONET; InferNetwork_ML_CV (gt0,gt1,gt2,gt3,gt4,gt5,gt6,gt7,gt8,gt9,gt10,gt11,gt12,gt13,gt14,gt15,gt16,gt17,gt18,gt19,gt20) 3 -x 5; END;
Command References
Y. Yu, J. Dong, K. Liu, and L. Nakhleh, Probabilistic inference of reticulate evolutionary histories, Under Review.