MPEx 3.2, with complete documentation, is now available!
Send Suggestions to
We encourage all MPEx users to use the version associated with the 'Start' button above. That version uses Java 7, which has the latest Java security features incorporated. We realize that some users may be constrained to using Java 6, and so we make an MPEx version available here for those people.
Welcome to Membrane Protein Explorer (MPEx), a tool for exploring the topology and other features of membrane proteins by means of hydropathy plots based upon thermodynamic and biological principles. For instructions, see Features at a Glance and Brief Instructions for MPEx below, or the full documentation. Current version (version history): 3.2.14. A useful adjunct to MPEx is MPtopo, a database of Membrane Protein topology accessible from MPEx.
MPEx is a Java program (not a Java applet) deployed using Java Web Start, which is part of the Java Runtime Environment (JRE). There has been recent concern over Java security vulnerabilities. These concerns are being addressed by Oracle, the provider of Java, but apply to Java applets running within web browsers and, as far as we can determine, do not apply to programs like MPEx. MPEx is a Java Web Start program that does not run within a web browser. Furthermore, MPEx is signed by a certificate issued by the certificate authority Thawte, Inc.
MPEx 3.2 requires version 7, or higher, of the JRE. Once the JRE is installed and MPEx is downloaded, you should not need to return to this page in order to run MPEx. You should be able to launch MPEx by using Web Start or from the Java control panel. Web Start will check automatically for new MPEx versions, and downloading an update should provide a dialogue to create a shortcut to the program on your desktop or in your Start menu. Mac users running Lion through Yosemite operating systems may want to look at Lion tips or Yosemite tips. If you have any problems (especially Mac users), please let us know, and send details of your system configuration by e-mail to . (Mac users, if you are running Yosemite, you need to be running at least Java build 1.8.0_40-b27 - you can check this with the Terminal command ‘java -version’).
You can read more about Java Web Start and Runtime Environment technologies here. If you do not have these technologies installed already, pressing the 'Start MPEx' button should give you the opportunity to do so. If you are a frequent user of MPEx, or other Web Start applications, you may find it useful to create a shortcut to Web Start on your system. See the Java Web Start page for more information.
We gratefully acknowledge the many ideas and suggestions of Dr. Alex Ladokhin that are incorporated into MPEx. We also are pleased to acknowledge the superb programming skills of Craig Snider, which have made this version of MPEx a reality. We thank Jung Kim for his help with the preparation of the MPEx documentation. Finally, it is a pleasure to thank Michael Myers for editing the documentation pages.
The authors and copyright holders of MPEx are Craig Snider, Sajith Jayasinghe, Kalina Hristova, and Stephen White. If you use MPEx results, please reference: Snider C, Jayasinghe S, Hristova K, & White SH (2009). MPEx: A tool for exploring membrane proteins. Protein Sci 18:2624-2628. [PubMed Link]. Please read the Legal Notices regarding the use of MPEx and MPtopo. We welcome comments and suggestions: Please send them to or For technical details regarding platform compatibility, etc., see About MPEx below.
MPEx is designed for examination of membrane protein sequences using hydropathy-plot methods popularized by Kyte and Doolittle (1982). That is, it is based on sliding-window analysis that represents an amino acid sequence as a sequence of numbers representing physical or statistical properties of the amino acid sequence. The most common physical property is hydropathy, which is based upon the free energy of partitioning of amino acids between water and membrane. This version of MPEx uses two types of hydropathy scales: Experiment-based whole-residue partitioning scales determined in this laboratory and experiment-based biological partitioning scales determined in collaboration with the von Heijne laboratory at Stockholm University. The whole-residue partitioning scales predict with considerable accuracy the transmembrane (TM) segments of membrane proteins of known structure as shown by Jayasinghe et al. (2001). The biological scale utilizes current knowledge of the code the Sec61 translocon to identify TM segments. The algorithm is the same as the one used by von Heijne laboratory on their ΔGpred server, but here we have implemented it as Java Web Start application.
In addition to these basic functions, MPEx can also be used to examine β-barrel membrane proteins using algorithms developed by Wimley. It also includes a totalizer module for estimating the total hydrophobicity and hydrophobic moment of amino acid sequences. An important feature of MPEx is that one can change the charge state of charged amino acids (Hydropathy Analysis only) and make amino acid substitutions. This feature allows one to develop hypotheses about how amino acid substitutions affect the stability of membrane and membrane-active proteins.
The MPEx window is divided into two major subwindows: By default, the left-hand plot panel shows the Hydropathy Analysis plot. Different analysis or data display modes can be selected with the tabs at the top of the plot panel. These include "Hydropathy Analysis", "Translocon TM Analysis", "β-Barrel Analysis", "Data Buffer Overlays", and "Totalizer". In the various 'analysis' modes, an analysis plot is calculated and displayed automatically when a sequence is entered. Each time a modification is made to the sequence, the plot is automatically recalculated.
The right-hand dark-blue control panel is for setting parameters used in computing or displaying a data plot. A window at the top of the control panel provides summary information, including (for hydropathy analysis) the hydrophobicity scale in use, residue at cursor, direction of partitioning, and free energy of transfer of the amino acids within the window. Similar windows may appear in this region for other analysis modes.
Just above the plot panel is the data buffer control bar, which permits plots to be stored in data storage "buffers", manipulated, and compared.
File loading, printing, and many other features are controlled through theappearing at the top of the window or screen.
The blue cursor in the plot panel allows the central residue of a sliding window to be identified. The residue at the cursor is indicated in a text box at the top of the Plot Panel. The cursor may be moved along the sequence either by using the left and right arrow keys or the mouse.
Residues of particular interest (e.g. aromatic) can be marked on the analysis plot. Use.
A user may mark or measure any region s/he wishes. Use.
All output is controlled by themenu. The whole window, including the control panel or just the plot panel, can be printed directly (set your printer to landscape orientation). The content of the Results Window can be saved as a plain ASCII text file. Analysis plot data can be saved in a tab-delimited format suitable for direct import into plotting/graphing packages such as Origin™ or Excel™. TM or loop sequence prediction data for the major analysis modes may also be saved.
Any analysis plot, including Control Panel settings, etc., can be stored in a buffer using the data buffer control panel above the plot region. Various plots stored in the buffers can be compared by selecting the Data Buffer Overlays tab at the top of the plot panel. Individual buffer contents can be loaded into the main workspace using the Load # menu on the data buffer control panel menu bar. Display characteristics of overlay data plots (e.g., color or layering order) can be modified from the parameter control panel menu on the right-hand side. Different data sets can be selected for comparison using the tabs at the bottom of the plot panel.
MPEx hydropathy analysis is based upon and incorporates principles of membrane protein stability determined in our laboratory [see White & Wimley (1999) Annu. Rev. Biophys. Biomolec. Struct. 28:319-365]. The experiment-based Wimley-White whole-residue hydrophobicity scales are used in the plots. The Octanol scale (Oct) measures a residue's free energy of transfer from water to the bilayer hydrocarbon core. For TM segments, the thermodynamic cost of transferring H-bonded peptide bonds (CONH) into the hydrocarbon core of the membrane is a major determinant of stability. We have recently shown that the Octanol scale describes well the total energetics of TM helix stability, including the cost of dehydrating the H-bonded peptide bonds. Tests with the MPtopo 3D_helix set of membrane proteins indicate that TM segments can be identified with high accuracy. See Jaysinghe et al. (2001) J. Mol. Biol. 312:927-934. MPEx, however, allows the CONH cost to be modified for illustrative purposes. The Interface scale (IF) measures a residue's free energy of transfer within an unfolded polypeptide chain from water to a phosphocholine bilayer interface. Both scales identify those regions of a peptide chain most likely to prefer association with membranes. The Oct-IF scale is useful because it identifies segments that tend to prefer a transbilayer helix conformation relative to an unfolded interfacial location. This is useful for examining the refolding of toxins, e.g. diphtheria toxins, on membranes.
We have constructed a database of membrane proteins of known topology (MPtopo), accessible from MPEx, that is useful for comparing the hydropathy behavior of proteins of unknown topology with the behavior of similar proteins of known topology.
The choices are Wimley-White Octanol, Interface, or Octanol-Interface whole-residue scales, as noted above. As discussed elsewhere ( whole-residue hydrophobicity scales), a major issue in all hydrophobicity scales is the exact cost of partitioning the H-bonded peptide bonds of alpha-helices into the membrane. Although we are now reasonably certain of this value (see above), MPEx allows the cost to be modified by means of the in the It consists of two buttons: raises the energetic cost of transferring the CONH groups into the membrane, making the transfer less favorable; lowers the cost, making transfer more favorable. The net effect is to shift the plot relative to the hydropathy ΔG = 0 line. The charge state of Asp, Glu, and His can be accounted for using Change Charge at Cursor.
Users may now define their own scale for hydropathy analysis by selecting the Other button under "Select Wimley-White Scale". The file chooser that appears displays an example, using the octanol scale, of the required syntax.
The effect of salt bridges on hydropathy plots can be investigated using the Set Salt Bridge feature. This is also available in the under .
Locate uses an algorithm based upon a 19 AA sliding window to select the regions most likely to be TM segments, based upon the selected hydrophobicity scale and other parameters (see below). MPEx ignores the window length parameter in the Locate mode. Scan produces a hydropathy plot using the selected sliding-window length and places red horizontal markers to indicate regions that favor transmembrane locations.
A sequence may be changed by selecting a residue with the cursor and using Change Residue To. This feature is especially useful for looking at the effect of residue mutations on hydropathy plots.
Selecting the Translocon TM Analysis tab at the top of the plot panel reconfigures MPEx to analyze transmembrane proteins based on translocon-mediated transmembrane helix assembly, considering amino acid position-dependent membrane insertion efficiency, as well as hydrophobic moment, transmembrane segment length, and flanking amino acid influences. The "full biological" hydrophobicity scales used are from the supplemental information in Hessa et al. 2007 ( Nature 450:1026-1030). More information on the translocon machinery is available here.
MPEx makes predictions of transmembrane sequence regions whose lengths correspond to a window size within the range selected on the Translocon TM Analysis control panel. Data for each window size are plotted in a different color, and the cursor can be attached to follow a particular data set.
Other controls are similar to those in the Hydropathy Analysis control panel. The button labeled Visit DGpred at Stockholm U. takes your browser to a tool that does similar server-based analysis at Stockholm University.
Selecting the β-Barrel Analysis tab at the top of the plot panel reconfigures MPEx to do a screening analysis for the identification of β-barrel membrane proteins. The analysis algorithm is based upon Wimley (2002), Protein Science 11:301-312. The screening algorithm uses amino acid composition and architecture of β-barrel membrane proteins of known structure to make predictions of TM β-strands, connecting β-hairpin loops, and the number of likely TM β-strands.
Scores are assigned for the likelihood of the entire sequence representing a β-barrel motif membrane protein (the β-barrel Sequence Score), the number of predicted β-strand peaks, and for selected windows representing two potential β-strand regions and the connecting β-hairpin loop. A β-barrel sequence score greater than 2 is a strong indicator that a sequence represents a β-barrel protein. These values are displayed in the summary window at the top of the Control Panel. The window scores and predicted TM and hairpin regions for the entire sequence, as well as any known TM regions, are displayed in the plot panel. See the full documentation for detailed information about MPEx.
Controls provided by the β-barrel control panel are similar to those in the Hydropathy Analysis control panel, including a summary window, the ability to send the cursor to a specific amino acid sequence position, visibility of the available data plot sets, setting the length of the transmembrane and hairpin window sizes, and changing and restoring residues within the sequence.
Selecting the Totalizer tab at the top of the plot panel reconfigures MPEx to use the Totalizer, a tool for calculating the hydropathy of peptides based on the Wimley-White scales. It allows the state of the N- and C-termini to be accounted for (acetylation, amidation, etc.).
The "full biological scale" used in Tranlocon TM Analysis can be selected in Totalizer, but doesn't take into account the terminal groups. User-defined hydropathy scales may also be selected.
Totalizer also produces helical wheel plots, calculates and marks the hydrophobic moment, and shows the direction of the moment. The wheel can be displayed from either the N-terminus or C-terminus perspective. It can also be rotated around its central axis and have its radius adjusted.
Selecting Batch Processing allows doing Hydropathy analysis or Translocon-TM analysis on multiple sequences without further user intervention beyond setting basic analysis parameters and selecting input sources and output data types and destinations. A file may contain multiple FASTA sequences.
An interface to the MPtopo database is accessible from . Proteins identified from a search can be loaded directly into MPEx.
Saving sessions. Your entire MPEx session can be saved and reloaded later in order to resume projects after an interruption. See and on the Toolbar menu.
When you select a sequence and enter or change your analysis parameters, the data analysis plot will be displayed automatically. The data analysis plot will be updated automatically whenever a parameter is changed.
The minimum requirement for the proper operation of MPEx is a protein sequence. You have several choices: Using the FASTA format; the first line beyond the ">" is optional). The sequence should be in the conventional UPPERCASE ONE LETTER CODE designations for the 20 common amino acids. All lowercase letters and other characters will be discarded by MPEx (d, e, and h are exceptions).option, you may (1) read in a file from the lab's MPtopo database of membrane proteins of known topology, (2) enter or paste a sequence of interest into a sequence input window, or (3) obtain a sequence from the Swiss-Prot database (use
The control panel on the right side of the main MPEx window or the menu allows you to select parameters that determine the analysis plot results (see below).
For hydropathy analysis, you may change/select the following parameters:
Pull-down menus from thealso provide several parameter selections for hydropathy analysis:
The hydropathy profile of the entered sequence will be plotted in the graph window. The black curve is the actual profile; the superimposed green curve is a smoothed version of the profile.
Regions that have a favorable free energy of transfer determined in Scan Mode are indicated with horizontal red bars. In Locate Mode, the red bars indicate MPEx's best guess about putative TM segments. These favorable regions, the AA sequences of these regions, and their total transfer free energies are listed in the Results Window. If MPtopo database files are being used, the known TM segments are marked by blue bars. User-designated TM segments are shown by rust-colored bars.
A hydrophobic moment plot for the current sequence can be displayed by selecting the HF Moment checkbox under Plot Visibility on the Control Panel.
The Residues in Window box on the Control Panel shows the amino acids contained within the sliding-window, centered on the residue at the cursor. If the residue at the cursor is D, E, or H, checking the Change Charge box on the Control Panel will change the charge state. In the Residues in Window box on the Control Panel, D, E, or H indicate charged, while d, e, or h indicate neutral residues.
As with Hydropathy Analysis, the parameter panel allows you to select parameters that determine the translocon transmembrane analysis plot results. The Translocon Transmembrane Analysis parameter panel allows selection of the following settings:
Several pull-down menus from the menu bar also affect Translocon TM analysis:
A hydropathy profile, based on the "full biological scale", is plotted for each window size within the selected range. Each window size will be represented by a different color, corresponding to the legend in the upper left of the plot region.
The cursor may be placed on any of the window plot curves with the control panel spinner marked Cursor on Data Set for Window. Window-size specific data in the control panel summary window will correspond to the data set the cursor is currently on.
Regions that have a favorable free energy of transfer are indicated with horizontal red bars. The lengths of these regions are constrained to correspond to the range of sliding window sizes selected. These favorable regions, the AA sequences of these regions, and their total transfer free energies are listed in the Results Window. If MPtopo database files are being used, the known TM segments are marked by blue bars. User-designated TM segments are shown by rust-colored bars.
A hydrophobic moment plot for the current sequence and data set the cursor is on can be displayed by selecting the HF Moment checkbox under Plot Visibility on the Control Panel.
The Residues in Window box on the Control Panel shows the amino acids contained within the sliding-window, centered on the residue at the cursor. These correspond to the data set the cursor is currently on.
As with Hydropathy Analysis, the parameter panel allows you to select parameters that determine the β-barrel analysis plot results. The β-barrel analysis parameter panel allows selection of the following settings:
Several pull-down menus from the menu bar also affect β-barrel analysis:
The Mark Residues feature allows you to identify "interesting" amino acids, including charged residues that are titratable near physiological pH (D, E, and H), non-titratable basic residues (K and R) that reveal topology via the so-called "positive inside rule," and aromatic residues (F, W, and Y) known to prefer membrane surface locations in membrane proteins of known 3D structure.
In β-barrel analysis, two plots can be displayed. β-strand scores are shown in violet, and β-hairpin scores are shown in tan. These plots can be selected individually on the control panel under Plot Visibility.
Regions of the β-strand score plot having y-axis values between 2 and 6 are indicated with horizontal red bars, the range in which most membrane-spanning β-strands are found.
Regions of the β-hairpin score plot having y-axis values greater than 6 are indicated with horizontal green bars. These segments are likely β-hairpin regions of the AA sequence.
As with hydropathy analysis results, β-barrel predicted transmembrane and hairpin regions of the AA sequence are listed in the Results Window. If MPtopo database files are being used, the known TM segments are marked by blue bars. User-designated segments are shown by rust-colored bars.
The Residues in Windows boxes on the Control Panel show the amino acids contained in the sliding window that represent a hairpin region of the AA sequence wherein the cursor is centered, flanked by two transmembrane regions.
For additional information, please refer to the complete documentation.
MPEx uses several public Java packages and classes that we are pleased to acknowledge.
MPEx is signed by a certificate issued by Thawte, Inc.
If you experience buggy behavior when using MPEx, or if certain functionalities are not working, please consult the list of known incompatibilities, system requirements, and tested platforms. We have made every effort to test MPEx for stability but cannot guarantee its performance on all platforms and browser combinations. We do not take responsibility for system crashes caused by running MPEx.