super5 command
Align sequences using the Super5 algorithm.
Examples
muscle -super5 seqs.fa -output aln.afa
muscle -super5 seqs.fa -output aln.afa -perturb 3 -perm abc
muscle -super5 seqs.fa -output replicate.@.afa -perturb 3 -perm all
muscle -super5 seqs.fa -output replicates.efa -perturb 3 -perm all
Description
The
super5 command uses the
Super5 algorithm to
align sequences. Input must be in
FASTA format.
By default, a single alignment is generated using default parameters and output is in aligned FASTA
format. Super5 is generally used for aligning large sets of sequences where the
PPP algorithm
(
align command) is too slow.
This command does not directly support generating complete ensembles, so the
-stratified,
-diversified and
-replicates options are not supported.
You can generate all
guide tree permutations
for a single
HMM perturbation by using
-perm all,
this creates four alignments, one for each variant of the guide tree. If
-perm all
is set, and the output filename contains
@, then alignments are
written to four different FASTA files where
@ is replaced
by is replaced by the replicate name, e.g.
abc.3. Otherwise,
all four alignments are written to one file in
ensemble FASTA (EFA) format.
To create a
stratified ensemble, run
super5
with
-perm all once for each perturbation seed and combine the outputs.
To create a
diversified ensemble, run
super5
once for each perturbation seed plus guide tree permutation and combine the outputs.
Multiple FASTA files can be combined into a single EFA file by using the
fa2efa command.
Options
-perturb SEED
Integer random number seed for generating
HMM perturbations.
Default SEED=0, which uses default HMM parameters.
Number of consistency iterations. Default 2.
-refineiters N
Number of refinement iterations. Default 100.
-nt
Input sequences use nucleotide alphabet (default auto-detect).
-amino
Input sequences use amino acid alphabet (default auto-detect).
-threads N
Number of threads. Default is the number of CPU cores, or 20
if the CPU has more than 20 cores.
