A Server for Protein Structure Modeling

Introduction

ModWeb is a server for comparative protein structure modeling. It depends on the large scale protein structure modeling pipeline, ModPipe, for its functionality. (Eswar et.al., Nucl. Acids. Res., 2003). The structural templates used to build models in ModPipe consist of a set of non-redundant chains extracted from structures in the PDB. Sequence-structure matches are established using multiple variations of sequence-sequence, profile-sequence, sequence-profile and profile-profile alignment methods. Significant alignments (E-value better than 1.0) covering at least 30 amino acid residues are selected for modeling. Models are built for each one of the sequence-structure matches using comparative modeling by satisfaction of spatial restraints as implemented in Modeller (Sali & Blundell, JMB, 1993). Finally, the resulting models are evaluated using several model assessment schemes and the best scoring models are returned to the user.

E-mail address

The e-mail address that is entered here is used by ModWeb to communicate with the user about completion of the calculation and the results of the calculation, when requested.

Run name

The run name entered here is used as a reference to communicate with the user by e-mail. This value is also used to name the dataset in ModBase, when requested.

Modeller access key

ModWeb is available free of charge for all academic non-profit institutions. Please read and accept the Academic License Agreement for Modeller to receive a key [Modeller Registration].

Input data

The input to ModWeb is one or more sequences in FASTA format. ModWeb will reject sequences which do not classify as FASTA. Each sequence entry in a typical FASTA file will contain a line beginning with ">" and then the name of the sequence. The actual amino acid sequence itself will start with the next line. An example of a FASTA formated sequence:

>sequence1
TLNGGITDMLTELANFEKNVSQAIHKYNAYRKAASVIAKYPHKIKSGAEAKKLPGVGTKIAEKIDEFLATGKL
RKLEKIRQDDTSSSINFLTRVSGIGPSAARKFVDEGIKTLEDLRKNEDKLNHHQRIGLKYFGDFEKRI
>sequence2
TSSSINFLTRVSGIGPSAARKFVDEGIKTLEDLRKNEDKLNHHQRIGLKYFGDFEKRIPREEMLQMQDIVLNEVKK
IAKYPHKIKSGAEAKKLPGVGTKIAEKIDE

Sequences pasted into the text box should be FASTA format also.

Selection of final models

Final models will be selected by one or more criteria specified in the form.

• The best scoring model as assessed by the normalized DOPE score (Shen & Sali, Prot. Sci., 2006)
• The longest model assessed good by normalized DOPE

In addition, models will also always be selected based on the following criteria:

• The model based on the alignment with the best sequence identity
• The best scoring model as assessed by the ModPipe quality score (MPQS)

Other options

The "search speed" option, despite its name, indirectly controls the amount of sampling done in the alignment space. It determines which methods are used to calculate sequence-structure alignments. "Very Fast" uses pairwise sequence alignments to calculate alignments; "Fast" uses sequence-sequence and sequence-profile methods; "Slow" uses sequence-sequence, profile-sequence and sequence-profile methods; and "Very Slow" includes all the of methods above and also adds profile-profile alignment methods.

The "Upload to ModBase" option will upload the calculated models to ModBase. If the input contains a single sequence the emails will also be emailed to you. But when the input contains more than one sequence, this option becomes mandatory.