Computational protocol: Phylogenetic Analysis of a ‘Jewel Orchid’ Genus Goodyera (Orchidaceae) Based on DNA Sequence Data from Nuclear and Plastid Regions

Similar protocols

Protocol publication

[…] Sequences were firstly assembled and edited with Seqman (DNA STAR package, Madison, WI, USA) [], aligned with Mega 5 [] and then adjusted manually. Three datasets, namely ITS, the combined chloroplast dataset (matK and trnL-F) and the combined nuclear and chloroplast DNA sequences (ITS, matK and trnL-F) were analysed using Bayesian inference (BI), maximum parsimony (MP) and maximum likelihood (ML); all characters were treated as unordered and equally weighted. Indels were treated as missing data.The maximum parsimony (MP) analyses were performed with PAUP* version 4.0b10 []. A heuristic search with 1000 random addition sequence replicates, tree bisection-reconnection (TBR) branch swapping and the MulTrees (saving multiple trees in memory) option were performed. Bootstrap values were generated with 1000 bootstrap replicates with TBR branch swapping, with each replicate performing 100 random-addition sequence replicates and a limit of 1000 trees. Homoplasy levels were assessed by means of the consistency index (CI) and the retention index (RI). For the ML analyses, MrModelTest 2.3 [] was used to select the most suitable model under the Akaike information criterion (AIC) []. The GTR+I+G model was selected as the best-fit model by MrMTgui 1.0 [] for all datasets. Then the models were added in a command block after the data in the NEXUS file. A total of 1000 bootstrap replicates were performed using Garli v0.951-GUI []. Other parameters were set as default for the Garli searches. PAUP * version 4.0b10 was used for exporting tree files. Both for the MP and ML analyses, bootstrap values over 89% were considered as high support, between 71% and 89% moderate support and below 71% as weak support. The Bayesian inference analyses were conducted with MrBayes 3.1.2 []. GTR+I+G was selected as the best-fit model as for the ML analyses for all datasets. The analyses consisted of 3,000,000 generations of four simultaneous Monte Carlo Markov chains. We increased the number of generations until the average deviation of split frequencies fell below 0.01 []. Trees were sampled every 1000 generations; the samples prior to stationary were discarded as burn-in using Tracer v. 1.5 and the remaining trees were used to build a majority-rule consensus tree on which the posterior probabilities (PP) were shown. We defined PP values above 0.90 as high support, between 0.80 and 0.89 as moderate support, and below 0.79 as weak support. […]

Pipeline specifications

Software tools MEGA, MrModelTest, GARLI, PAUP*, MrBayes
Application Phylogenetics