cipres_submit_raxml: Submit RAxML analysis
In fmichonneau/cipresr: An Interface to the 'CIPRES REST API' (CRA)

Description Usage Arguments Details Author(s)

Submit to CIPRES an RAxML job. For more information, please consult the RAxML manual available on the RAxML website (http://sco.h-its.org/exelixis/web/software/raxml).

cipres_submit_raxml(input_file, select_analysis = c("fa", "fd", "fo", "fA",
  "J", "y", "I"), starting_tree = NULL, constraint = NULL,
  binary_backbone = NULL, partition = NULL, exclude = NULL,
  weights = NULL, user_prot_matrix = NULL, secondary_structure = NULL,
  bootstrap_topologies = NULL, max_runtime = 10, CAT_categories = NULL,
  parsimony_seed = NULL, datatype = c("dna", "protein", "rna", "binary",
  "mutli"), outgroups = NULL, p_inv = FALSE, empirical_freq = TRUE,
  print_brlen = FALSE, more_memory = FALSE, n_char = NULL, n_tax = NULL,
  disable_check_sequence = FALSE, mesquite_output = FALSE,
  model = "GTRGAMMA", n_runs = 1, no_bfgs = FALSE,
  intermediate_tree_files = FALSE, convergence_criterion = FALSE,
  majority_rule = c("MRE", "STRICT", "MR_DROP", "STRICT_DROP"),
  protein_matrix = c("DAYHOFF", "DCMUT", "JTT", "MTREV", "WAG", "RTREV",
  "CPREV", "VT", "BLOSUM62", "MTMAM", "LG", "MTART", "MTZOA", "PMB", "HIVB",
  "HIVW", "JTTDCMUT", "FLU", "DUMMY", "DUMMY2", "AUTO", "LG4M", "LG4X",
  "PROT_FILE", "GTR_UNLINKED", "GTR"), bootstrap_type = c("b", "x"),
  bs_seed_value = 12345, n_bootstrap_rep = NULL,
  bootstop_type = c("autoMRE", "autoFC", "autoMR", "autoMRE_IGN"),
  job_name = NULL, get_email = TRUE, ...)

`input_file`	An alignment file in relaxed interleaved for sequential PHYLIP (or FASTA) format (-s)
`select_analysis`	The type of analysis to be performed.
`starting_tree`	A NEWICK tree file specifying the starting tree. If `NULL` (default), RAxML uses a randomized stepwise addition Parsimony starting tree.
`constraint`	A NEWICK tree file specifying topology constraining the tree search. If `NULL` (default), no constraint is used.
`binary_backbone`	The file name of a binary constraint tree. If `NULL` (default), no constraint is used.
`partition`	A file specifying the partitions in the alignment file using the RAxML format (refer to the RAxML manual for details). If `NULL` (default), the full alignment is considered as a single partition.
`exclude`	A file specifying positions in the alignment to exclude. If `NULL` (default), no position is excluded.
`weights`	A file indicating the weights for each column position in the alignment. If `NULL` (default), all alignments positions have identical weights.
`user_prot_matrix`	A file containing an amino-acid substitution model.
`secondary_structure`	A file specifying a secondary structure model for the alignment.
`bootstrap_topologies`	A file containing topologies for a posteriori bootstrapping.
`max_runtime`	How long should the job run for? (in hours)
`CAT_categories`	When using a CAT model, how many rate categories (-c)?
`parsimony_seed`	For all options that use randomization, this seed must be specified (-p).
`datatype`	The type of data used in the alignment.
`outgroups`	Specify in name of a single outgroup (“Cat”) or a comma-separated list of outgroups (no spaces: “Cat,Dog,Bird”). If several species are listed as outgroup, only the first one in the list will be used a outgroup.
`p_inv`	Should a proportion of invariant site be estimated? (default `FALSE`).
`empirical_freq`	Should empirical frequencies be used with amino-acid alignments? (default `FALSE`).
`print_brlen`	Should branch lengths be printed on the screen? Default `FALSE`.
`more_memory`	Set to `TRUE` if your analysis requires more than 125GB of memory.
`n_char`	If using `more_memory`, set to the number of characters in your alignment to estimate the amount of memory needed.
`n_tax`	If using `more_memory`, set to the number of taxa in your alignment to estimate the amount of memory needed.
`disable_check_sequence`	Disable checking for completely undetermined sequences (it normally doesn't make sense to incude them in the analysis). Default `FALSE`. (`-O`).
`mesquite_output`	Generate output files that can be parsed by MESQUITE. Default `FALSE`. (Not allowed by CIPRES in runs with bootstrapping or alternative runs are used) (`-mesquite`).
`model`	The substitution model (`-m`).
`n_runs`	The number of alternative runs on distinct starting trees (`-#` or `-N`).
`no_bfgs`	If `TRUE`, disable the BFGS method when optimizing the GTR rates (default `FALSE`).
`intermediate_tree_files`	If `TRUE`, the intermediate tree will be written to files (`-j`).
`convergence_criterion`	ML search convergence criterion uses Robinson-Foulds distance. Default `FALSE`. Use on trees with large number of taxa (`-D`).
`majority_rule`	When computing the majority rule consensus tree (`select_analysis="J"`), which method to use?
`protein_matrix`	Which amino-acid substitution matrix to use when working with protein sequence alignment?
`bootstrap_type`	Which type of bootstrapping algorithm to use (with `select_analysis` is `fa`, `fd`, or `fo`)? `x` for rapid bootstrapping and `b` for normal bootstrapping.
`bs_seed_value`	The seed to use for the bootstrapping.
`n_bootstrap_rep`	If `NULL`, the number of bootstrap replicate will be selected by RAxML according to the algorithm specified by the option `bootstop_type`. Other an integer specifiying the number of replicates.
`bootstop_type`	The type of algorithm used to estimate the number of replicates.
`job_name`	An optional name to identify the job.
`get_email`	Should an email be sent when the job completed?
`...`	arguments to customize the API call. These arguments are passed to the appropriate `httr` function (`GET`, `POST`, `DELETE`)

CIPRES offers several flavors of RAxML but cipresr only provides an interface with the full RAxML version. Setting the correct options for a given run can be challenging as RAxML is complex and feature-rich software. Read the RAxML manual carefully, double check for potential warnings and error messages in the output files, and your results.

Francois Michonneau

fmichonneau/cipresr documentation built on May 16, 2019, 1:44 p.m.