Structure of an amino acid sequence prediction (This Document is currently in progress)
The structure and function prediction pipeline is designed for building protein structures and predicting their functions using popular threading tools in association with Spanner and SeSAW.
Input your sequence with or without a FASTA-formatted header in the Query Sequence textbox. There are two actions that can be selected at this stage. If you are unsure about domain boundaries, then you can perform a 2D prediction, which gives secondary structure and intrinsic disorder propensity. If you are sure about domain boundaries, then you can go directly to 3D prediction. When you are satisfied with the input, click on the Run button. If your query is launched with success you will be brought to the progress tab.
This screen will show you the progress/status of your jobs. It will allow you to cancel jobs in progress and view the results of your finished job. There are various reasons why your job might fail. If you are confident that your input was proper and that an internal error likely occurred, please send use your feedback.
This screen shows prediction of intrinsic disorder propensity and secondary structure. Based on this information, you can easily select subsequences for structure modeling (Spanner) or, if the sub-sequence is predicted to be disordered, composition-based function prediction (IDD Navigator). Selections can be made in one of several ways.
You can use the Quick Selection bar to define mostly disordered (pink) or mostly ordered (purple) regions with one click. You can left-click and drag your mouse over the region of interest. The selection can then be fine-tuned by dragging the edges of the selected region left or right. You can enter the Range of your selection directly Note that the Overview panel always shows the full-length sequence as well as the selected region
This screen will display the results of threading programs. Each result line, ranked by the threading program, contains basic information, such as the PDB ID of the template, score, and links to graphical views of the template. Clicking on a result shows more information about the alignment. Additionally you the option to run the selected alignment with Spanner to build a 3D model.
This screen displays the constructed model of the query based on the selected alignment. You will see a static rendering of the model, as well as a button to switch between the static view and an interactive view of the protein. Note that the alignment may be slightly modified if the template coordinates in the original alignment are not defined in the actual PDB file. Links to download the input and output files of spanner are also available.
[Work in progress]
The results will be kept for 72h, along with your input data.
The HH-suite is an open-source software package for highly sensitive sequence searching and sequence alignment. Its two most important programs are HHsearch and HHblits. Both are based on the pairwise comparison of profile hidden Markov models (HMMs).
A profile-profile alignment utilizes information present in sequences of homologous proteins to amplify the sequence conservation pattern defining the protein family. This method allows detection of remote homologies beyond the reach of other sequence comparison methods. From the sequence, a profile is generated that is then compared to several databases of sequence profiles of proteins.
A profile-profile comparison tool for protein fold recognition.
Basic Local Alignment Search Tool, and its Position-Specific Iterated variant.
Spanner is a structural modeling program that threads a query amino-acid sequence onto a template protein structure. Spanner handles gaps by spanning the region of interest using fragments of known structures. Sidechains are replaced using one of several methods, and the final model is minimized by either the GROMACS or MyPresto molecular mechanics programs.
Sesaw is a service for locating sequentially and structurally similar motifs to a query structure by ASH alignment, followed by scoring with position-specific scoring matrices (PSSMs).
Jmol is a free, open source molecule viewer for students, educators, and researchers in chemistry and biochemistry. It is cross-platform, running on Windows, Mac OS X, and Linux/Unix systems.