Protter

Welcome to the user manual of Protter - a webservice and webapplication to visualize the sequence, topology and annotations of individual proteins.

You can access the interactive webclient at https://wlab.ethz.ch/protter to create protein plots.

Alternatively you can programmatically use the webservice at https://wlab.ethz.ch/protter/create using a set of defined parameters.

You can also download a pdf-version of this user manual here.


If you have additional questions, encounter any error or bug, or have suggestions on how to improve Protter, please contribute to the Protter Discussion Group.


More documentation on:

Web Application

Using the Protter web application you can easily create high-quality protein topology plots with computed, curated and custom protein features annotated in a highly interactive way.

This is how the web interface looks like: In the upper region, there are four tabs to specify various settings. While in the lower region, there is the actual protein visualization. Whenever you change a parameter, the visualization will automatically refresh and show your changes instantaneously.

Typically, you will go through the following steps as indicated at the top of the screen:

  1. Protein

    You start at the "protein" settings where you specify the protein that you want to visualize. Just enter the UniProt protein identifier of your protein. You can also provide gene names and Protter will ask you back in case that this gene name is ambiguous within UniProt.

    • Click  submit to set the identifier and create a first visualization.
    • The  load example button will enter the identifier BST2_HUMAN for you.
    • The  enter a list of proteins and the  load a proteomics result file button allow you to specify multiple proteins for batch visualization. See the help section "Web Application - Batch Processing" for details.

    In case you want to visualize an arbitrary amino acid sequence, simply switch to "by sequence" and enter your sequence in the textbox:

    • Click  submit to set the sequence and create a first visualization.
    • The  load example button will enter an exemplary amino acid sequence for you.
    • The  load FASTA file button allows you to load a protein sequence file in FASTA format. See the help section "Web Application - Batch Processing" for details.

  2. Topology

    Specification of the protein's sequence has generated a first visualization using default parameters. In the "topology" settings, you can fine tune the protein's topology.

    A protein's transmembrane topology consists of: N-terminus location, transmembrane regions and, optional, membrane anchors:

    In the first panel, you can switch between "no membrane", "automatic" (default) and "custom".

    • "no membrane" will completely hide the membranes, which is useful for non-membrane proteins.
    • "automatic" will come up with a transmembrane topology depending on your input sequence: for UniProt identifiers it will use the annotated topology, and for amino acid sequences it will use a transmembrane topology predicted by Phobius.
    • "custom" allows you to specify all the details yourself:
      • "N-terminus location" — sets the N-terminus be located outside, inside or as annotated by UniProt or as predicted by Phobius.
      • "transmembrane regions" — set the transmembrane regions by adding an arbitrary number of regions, see also Definition of the regions. You can interactively select from a pre-selected set of annotated regions by clicking  in the textfield.
      • "membrane anchors" — set the membrane anchor regions by adding an arbitrary number of regions, see also Definition of the regions. You can interactively select from a pre-selected set of annotated regions by clicking  in the textfield.

  3. Styles

    In Protter, Styles are used to highlight certain regions of the protein sequence by varying four parameters:

    1. the color of the amino acid letter
    2. the shape of the amino acid symbol
    3. the symbol’s frame color
    4. the symbol’s background color

    The "styles" settings shows a table of all currently specified styles. Each row represents a style and has 8 columns:

    • "name" [optional] — represents an arbitrary name for the style. This name will also be used for the legend.
    • "shape" — set the shape of the amino acid symbol. Use the  button to not set a shape for this style.
    • "char color" — set the color of the amino acid letter. Click on the color-square to open up a color selection window: Click on the desired color or choose the  no change option to not set a char color for this style.
    • "frame color" — set the frame color of the amino acid symbol.
    • "back color" — set the background color of the amino acid symbol.
    • "region" — in this textfield enter one or more regions of your protein that this style should be applied to. Multiple regions can be separated by new lines or by commas.
      A region can be:
      • a single position, e.g.: 5
      • a range of positions, e.g.: 123-144
      • a plain sequence, e.g.: ELVISISALIVE
      • a regular expression matching parts of the sequence. If there is a regexp-group specified, only the group of the match will be part of the region, e.g.: (N).[ST]
      • an annotated region. You can interactively select from all available annotated regions by clicking  in the lower-right part of the textfield:
    • "active" — unselect this toggle-button to exclude this style from visualization but still keep it in the style table.
    • "delete" — click to remove this style from the visualization and the style table.

    If highlighted regions (partly) overlap, styles defined further down will override styles defined further up. You can drag-and-drop the rows in the table using the double-arrow  on the very beginning of each row to reorder the priority of your styles.

    Click  at the end of the style table to add a new empty style.

    Click  at the end of the style table to remove all styles at once.

  4. Miscellaneous

    In the "misc." settings you can modify details of the protein visualization.

    • membrane color — set the fill color of the membrane.
    • TM label color — set the color of the transmembrane-domain labels.
    • TM label style — set the style of the transmembrane-domain labels. Select none to not show labels for the transmembrane-domains.
    • protease — select a protease to display protease specific cleavage sites as dashed lines in your visualization.
    • label every 10th residue — add small position labels to your protein sequence visualization.
    • show legend — add a legend summarizing all the styles used.
    • TeXtopo options — specify additional options for the underlying TeXtopo layout generation (e.g. flipNterm, loopextent, loopfoot, etc; see TeXtopo documentation for a complete list of options).

Visualization

Move the mouse cursor on any amino acid to show its position within the protein and the styles applied to this position. To interactively select a range, click the amino acid at the beginning of the desired range and subsequently click the amino acid at the end of the desired selection. After selection the range will be available in the annotation dropdown menu for styling regions.

Visualization settings

In the lower region of the web application, next to the protein visualization, there are some settings to adjust the visualization itself:
 zoom in to the protein
 zoom out of the protein
 set zoom to fit whole protein
 refresh the protein visualization
 download protein visualization as PDF, SVG, or PNG file
 send a link to the current Protter visualization via email
 open currently visualized protein in UniProt (for proteins entered "by accession") or BLAST currently visualized protein in UniProt (for proteins entered "by sequence").

Web Application - Batch Processing

On top of visualizing your protein of interest you can also load a list of proteins, browse through the visualizations, and download a combined visualization of all proteins as one PDF document or as ZIP archive of separate files. This is particularly useful to visualize results of mass-spectrometry based proteomics studies.

Loading Proteins and Peptides

There are different ways of loading a list of proteins in Protter:

Browsing Proteins and Peptides

After having loaded a FASTA file, a list of protein IDs or a proteomics result file, Protter will insert a protein/peptide browser panel on the left side with a tree-like view of all proteins and peptides (optional):

By clicking a protein, Protter will immediately load the selected protein in the visualization pane. Additionally, if specified, the list of peptides of this protein will be shown.

Clicking a peptide sequence will select (see Visualization section) the corresponding region on the amino acid sequence in the protein visualization:

Click  reset to unload all proteins/peptides and reset Protter to start over again.

Click  export all... to open up the export all proteins dialog:

Exporting visualizations for all proteins

Click  export all... at the bottom of the protein table to open up the export all proteins dialog: The protein visualizations can be exported to: The process of batch export can take some minutes depending on the number of proteins. You will be notified by email when the results are ready for download.

Note: the public Protter server is currently limited to 100 proteins in batch export. To generate visualizations of bigger protein lists you can download and setup your own local Protter server. See Installation & Requirements of the Server software section below.

Parameters for API access

All parameters have to be separated by a &
e.g.: create?param1&param2&param3&...

seq | up

To specify a sequence you can either enter the sequence itself via the seq parameter
e.g.: seq=ELVISISALIVE

or a Uniprot Name or Accession by the up parameter.
e.g.: up=CD44_HUMAN

Note: only one of both should be set.

tm

tm is used to set the transmembrane regions. This parameter is optional and will result in a visualization without a membrane if omitted.
e.g.: tm=phobius.tm
Note: if tm=auto transmembrane topology will be automatically determined, which is equivalent to nterm=UP.NTERM&tm=UP.TRANSMEM&anchor=UP.LIPID for a UniProt identifier and nterm=PHOBIUS.NTERM&tm=PHOBIUS.TM for any other given sequence.

nterm

nterm specifies the location of the protein's N-terminus. It can be intra | extra | up.nterm | phobius.nterm. This parameter is optional and will result in a visualization without a membrane if omitted.
e.g.: nterm=intra

anchor

anchor specifies regions of membrane anchors which will be considered in the visualisation of the protein's topology.
e.g.: anchor=UP.LIPID

format

specifies the requested image format. Currently supported formats are svg | png | pdf.
e.g.: format=svg

mc

mc sets the membrane color.
e.g.: mc=lightsalmon

lc

lc sets the transmembrane label color.
e.g.: lc=blue

tml

tml sets the transmembrane label style and can be either none | numcount | alphacount | Alphacount | romancount | Romancount.
e.g.: tml=Alphacount
Note: the value is case sensitive!

cutAt

specifies the positions where potential enzymatic cleavage should be indicated. This parameter is usually used with a peptidecutter-region.
e.g.: cutAt=peptidecutter.Tryps

lblin | lblout

override the default labels at the membrane (intra & extra).
e.g.: lblin=Nuclear&lblout=Perinuclear%20space

numbers

add this parameter to get number-labels at every 10th residue.
e.g.: numbers
Note: this parameter does not need a value.

title

adds a centered title to the top of the protein plot.
e.g.: title=MyProtein

legend

add this parameter to include a legend of all defined styles right next to the protein plot
e.g.: legend
Note: this parameter does not need a value.

Definition of the styles

A style is defined and applied to a collection of regions as a single parameter:
e.g.: style=regionA[,regionB,...]
Numerous styles may be displayed by treating them as individual parameters (i.e. separated by a &).

A style is made up of one or more comma separated style-parameters: p1:v1[,p2:v2,...]
Later style definitions will override previous ones and if a style-parameter is omitted the value of the previous style will be used.
Following style-parameters are available:

e.g.: s:circle,fc:black,cc:black,bc:white
Note: All color names have to be standard SVG colors!

Definition of the regions

Regions are a collection of one or more individual regions. The individual regions are to be separated by commas.
e.g.: regionA[,regionB,...]

An individual region can be either:

Phobius predicitons

When specifying a Phobius region, the protein sequence will be sent to a Phobius server for topology prediction. The URL of the Phobius server can be set in the configuration file. Following Phobius prediction annotations are supported:

Uniprot annotations

Uniprot provides numerous annotations for its proteins. See http://www.uniprot.org/manual/sequence_annotation for all available Uniprot sequence annotations including descriptions and examples.

PeptideCutter cleavage sites

PeptideCutter provides a webservice to in-silico digest protein sequences. PeptideCutter regions will refer to the amino acids immediately preceding the reported cleavage sites.

Regular Expressions

Regular Expressions are a mighty concept for defining protein sequence motifs. To build regular expressions with live feedback http://www.gethifi.com/tools/regex provides a helpful tool.



Note: The protter web application may be called using the same parameters as the ProtterServer: https://wlab.ethz.ch/protter/#param1&param2&param3&...
This urls allow interactive editing of the pre-coded visualization and can be readily bookmarked or send via email.

Installation & Requirements of the Server software

To install your own local copy of Protter, please follow these steps:

Implementation details

Protter Server

The ProtterServer is a webservice to visualize the sequence, topology and annotations of individual proteins. The server is written in Java. The webserver functionality is based on NanoHTTPD (version 1.21). For visualization a request is transformed into a TeX file which is processed within a LaTeX distribution (e.g. MiKTeX) by the TeXtopo package. Textopo takes care of the layout of protein topology. The resulting dvi file is converted to a svg file using dvisvgm. The initial svg is further processed to include the specified styles. Optionally, the resulting svg may be converted to a png or pdf image using the Apache Batik toolkit. To allow for visualization of annotations ProtterServer is using several other webservices to resolve sequence specific protein annotations. Currently UniProt, Phobius and PeptideCutter can provide sequences and annotations.

Protter WebApp

The Protter web application is a browser based interactive client for the Protter server. It is based on the AJAX framework jQuery (version 1.9, http://jquery.com/) extended by following plugins:

Annotation Data Sources

Protter not only displays your custom protein annotation but also retrieves and displays computed/curated annotations from various sources:

Colors

Various parameters require the specification of a color. You can choose from 140 SVG compatible colors (see below) or specify any color by a hexadecimal code (e.g. FF0000 for red).

 aliceblue  antiquewhite  aqua  aquamarine  azure  beige
 bisque  black  blanchedalmond  blue  blueviolet  brown
 burlywood  cadetblue  chartreuse  chocolate  coral  cornflowerblue
 cornsilk  crimson  cyan  darkblue  darkcyan  darkgoldenrod
 darkgray  darkgreen  darkgrey  darkkhaki  darkmagenta  darkolivegreen
 darkorange  darkorchid  darkred  darksalmon  darkseagreen  darkslateblue
 darkslategray  darkslategrey  darkturquoise  darkviolet  deeppink  deepskyblue
 dimgray  dimgrey  dodgerblue  firebrick  floralwhite  forestgreen
 fuchsia  gainsboro  ghostwhite  gold  goldenrod  gray
 green  greenyellow  grey  honeydew  hotpink  indianred
 indigo  ivory  khaki  lavender  lavenderblush  lawngreen
 lemonchiffon  lightblue  lightcoral  lightcyan  lightgoldenrodyellow  lightgray
 lightgreen  lightgrey  lightpink  lightsalmon  lightseagreen  lightskyblue
 lightslategray  lightslategrey  lightsteelblue  lightyellow  lime  limegreen
 linen  magenta  maroon  mediumaquamarine  mediumblue  mediumorchid
 mediumpurple  mediumseagreen  mediumslateblue  mediumspringgreen  mediumturquoise  mediumvioletred
 midnightblue  mintcream  mistyrose  moccasin  navajowhite  navy
 oldlace  olive  olivedrab  orange  orangered  orchid
 palegoldenrod  palegreen  paleturquoise  palevioletred  papayawhip  peachpuff
 peru  pink  plum  powderblue  purple  red
 rosybrown  royalblue  saddlebrown  salmon  sandybrown  seagreen
 seashell  sienna  silver  skyblue  slateblue  slategray
 slategrey  snow  springgreen  steelblue  tan  teal
 thistle  tomato  turquoise  violet  wheat  white
 whitesmoke  yellow  yellowgreen


imsb.biol.ethz.ch
Ulrich Omasits - protter@