NAME

       hmm_train  -  Estimate  the  transition  probabilities  of  an  HMM,  based on multiple -g
       <gff_fname_list> [OPTIONS] > out.hmm

DESCRIPTION:

       Estimate the transition probabilities of an HMM, based on  multiple  alignments,  sequence
       annotations, and a category map.

OPTIONS

   required options
       -m <msa_fname_list>

              List  of  multiple  sequence alignment files.  Currently, in testing mode, the list
              must be of length one.

       -c <category_map_fname>

              File defining mapping of feature types to category numbers.

       -g <gff_fname_list>

              Files in GFF defining sequence features to be used in labeling  sites.    Frame  of
              reference of feature indices is determined feature-by-feature according to

       'seqname' attribute.
              Filenames must correspond in number and order

              to the elements of <msa_fname_list>.

   alignment options
       -M <msa_length_list>

              (Mutually   exclusive   with   -m)  Assume  alignments  of  the  specified  lengths
              (comma-separated list) and do not  not  attempt  to  map  the  coordinates  in  the
              specified  GFFs  (assume  they  are  in the desired coordinate frame).  This option
              allows an HMM to be trained directly from GFFs, without alignments.  Not  permitted
              with -I.

       -i PHYLIP|FASTA|MPM|SS

              (default SS) Alignment format.

       -R <tag>

              Before   estimating  transition  probabilities,  group  features  by  <tag>  (e.g.,
              "transcript_id" or "exon_id") and reverse  complement  segments  of  the  alignment
              corresponding to groups on the reverse strand.  Groups must be non-overlapping (see
              refeature --unique).

   indel options
       -I <indel_cat_list>

       Model indels for specified categories.
              To have

              nonzero probability for the states corresponding to  a  specified  category  range,
              indels  must  be  "clean"  (nonoverlapping),  must  be assignable by parsimony to a
              single branch in the phylogenetic tree,  and  must  have  lengths  that  are  exact
              multiples  of  the  category  range  size.   Avoid -G with this option.  If used in
              training mode, requires -T.

       -t <tree_fname>

              Use the specified tree topology when training for indels.

       -n <nseqs>

              Train an indel model for <nseqs> sequences, despite that the training alignment has
              a  different  number.   All  (non-trivial)  gap  patterns are assumed to be equally
              frequent.

   other options
       -q

              Proceed quietly (without updates to stderr).

       -h

              Print this help message and exit.