About HMMSTR
==== HMMSTR says ======
============ THE SERVER IS BROKEN =================
Sorry. Please come back later.
Rosetta is provided as-is. This is a 1998 version. If you want a
much higher quality prediction, try
ROBETTA
Local structure predictions (HMMSTR-CM, I-sites) are still up and running.
to www.bioinfo.rpi.edu/bystrc/hmmstr/server.php
PLEASE NOTE: the new URL is minus the tilde '~'. If you have made a bookmark, update it.
Time to fix bugs is scarce. Please be patient...
Announcing I-sites2 fragment predictions, using PAIRWISE SEQUENCE COVARIANCE!!
...and introducing a nice clickable image of the Isites predictions.
Quota in effect. Please submit no more than 24 jobs per day.
<--->
[
Wed Jan 26 22:40:30 EST 2011
]
Email address to which results are to be sent:
Check your spelling! If the address is wrong, you will receive
no messages!
Add a short identifying remark for this sequence:
Type or paste your amino acid sequence information below.
SPECIFY INPUT FORMAT:
Single sequence:
1-letter code.
from PDB coordinates.
Multiple sequence alignment file:
HSSP
FASTA
Clustal
SAF
format:
Optional chain identifier:
Input options:
Use
PSI-BLAST to generate multiple sequence alignment.
Use a
low-complexity filter.
Output options:
A multiple sequence alignment in SAF format (only if using PSI-BLAST).
example
Sequence names and weights used for generating the sequence profile.
example
A set of multiple, predicted fragments sorted by confidence.
example
A set of predicted backbone angles and confidences based on linking fragments.
example
A one-dimensional prediction of backbone angle (phi/psi) regions based on linked fragments.
example
A prediction of 3-state secondary structure, backbone angles and context symbols using
HMMSTR.
example
A contact map prediction using HMMSTR-CM
example
Complete list of HMMSTR state probabilities
example
A set of 3D coordinates in PDB format, generated using ROSETTA.
example
===========EXPERT SETTINGS===========
I-sites method to use:
I-sites v.1
PHD + I-sites1
(example)
I-sites v.2
HMMSTR model to use:
R
D
C
Maximum e-value for PSI-BLAST:
Ignore I-sites shorter than
in fragment prediction report.
Minimum allowed percent identity for inclusion in profile :
Minimum allowed confidence of fragment prediction:
Number of parent sequence in an input multiple alignment (usually 1) :
Create an unrefined homology-based model if possible:
Label for internal use:
Number output coordinates starting from:
=====================================