Abie Pro 3.0 Demo Help: Peptide Antibody Design

Why Abie Pro? Online

Abie is specifically written for peptide antibody design. It has a friendly user interface for selecting a peptide antigen based on peptide length, hydrophicility, amino acid preference, and post-translational modification (see below).

One significant problem in antibody production is the background cross-reactivity. Blast search may eliminate any peptide candidate that shares sequence similarities with many unwanted proteins. Blast sequence similarity search, however, will miss many cross-reactive proteins since an epitope --- amino acid sequences recognized by a specific antibody --- could be as short as 4 - 6 amino acids. Blast is designed for comparing sequences in a longer range.

Abie Pro analyzed (in advance) known protein sequences in a proteome for the frequency of short peptide sequences. For an example, Abie searched a non-redundant mouse protein database and found that the quadri-peptide SGSG is present in the mouse proteome for 5197 times in 3170 proteins, and the quadri-peptide ACWM is present for only 2 times in 2 proteins. SGSG is over-represented in the mouse proteome. If your peptide happened to contain an over-represented sequence, it is likely a polyclonal antibody raised against the peptide will have a fraction of antibodies that reacts with the over-represented sequences, therefore cross-reacts with a large number of unwanted proteins. Because over-represented sequences are over-represented, there is a non-negligible probability that your peptide will contain one over-represented short sequence. Abie Pro 3.0 has an option to eliminate any peptide containing over-represented sequences, as a result, may reduce background cross-reactions.

Figure 1. The frequency of quadri-peptides in a mouse proteome. A small fraction of short amino acid sequences are "over-represented" in a proteome. The line is truncated at the end, i.e., there is a large number of "under-represented" sequences not shown.

A lot of calculations have been done in advance for Abie Pro. That is why Abie Pro runs amazingly fast in your web browser. But to take full advantage of Abie Pro, you would need to install it on your computer. For example, the online Abie does not allow you to chose a species other than mouse because limitations in passing large number of data on the web. Note each species has its unique proteome, the proteome data for the species of your target protein (or the closest relative) should be used whenever it is possible. The stand-only version of Abie Pro 3.0 has proteome data for 21 species. Click here for ordering information.

Browser requirements:

Abie Pro 3.0 has a stand-only version and an online version. The stand-only version installs on user's computer, it does not require any web browser to run. It also runs faster, has fewer bugs and more functions than the online version.

The Abie Pro 3.0 Online you see here runs on the following Java-enabled browser:

PC: Internet Explorer 5.5 or higher and Netscape 4.08 or higher.
Mac (OS 9.*): Internet Explorer 5.0

If you use one of the above browser and you can't run Abie Pro 3.0 Online, please make sure Java is enabled in your browser. For Internet Explorer, go to Tools/Internet Options, click on Security Settings, scroll down to find Microsoft VM, deselect "Disable Java".

Mac OSX 10 IE5.1 has a bug, it does not allow copy/paste into a Java text field, thus you will not be able to import a new sequence. Attempting copy/paste may kill your browser. Later Mac IE versions should work fine.

How to change parameters:

  • 1. To import a new sequence, copy and paste the sequence into the input window.
  • 2. To change the hydrophilicity threshold, click on the top half of the hydrophilicity graph.
  • 3. To display protein sequences, click on the bottom half on the hydrophilicity graph or on a sequence in the result window.
  • 4. To select amino acids that you don't want to include in your peptide, click on the amino acid name in the pull down menu. To select multiple choices, you may need to hold down the shift or control key on a PC. Click again, should deselect. On Mac, you may need to hold down the apple key and click to deselect.
  • 5. Similarly, you can select a subset of post-translation modifications you would like to avoid.
  • 6. Check the "Avoid background cross-reactions" will eliminate peptides with over-represented sequences. There are two parameters for this function: (1) Species --- choose the species of your target protein or a close relative. If the species name of your target protein is not present in the pull-down list and a close relative can't be found, uncheck "Avoid background cross-reactions." (2) Specificity --- "Low" for most stringent definition of "over-representation" and "High" for less stringent definition of "over-representation." In another word, if one choses "High" specificity, a peptide will have a higher probability of containing a "over-represented" sequence, thus more candidate peptides will be eliminated.

    Species available in the stand-only version

    Species Name Common Name
    Arabidopsis thaliana Arabidopsis
    Bacillus subtilis B subtilis
    Saccharomyces cerevisiae Budding yeast
    Caenorhabditis elegans C elegans
    Gallus gallus Chicken
    Zea mays Corn
    Bos taurus Cow
    Dictyostelium discoideum Dictyostelium
    Drosophila melanogaster Drosophila
    Escherichia coli E Coli
    Schizosaccharomyces pombe Fission yeast
    Xenopus laevis Frog
    Helicobacter pylori H pylori
    Human immunodeficiency virus type 1 HIV
    Homo sapiens Human
    Mycobacterium tuberculosis M Tuberculosis
    Mus musculus Mouse
    Sus scrofa Pig
    Rattus norvegicus Rat
    Oryza sativa Rice
    Danio rerio Zebrafish

    Contact us if a species you are interested in is not on this list, we might be able to help.

    Behind the scene:

    Several rules are now accepted for selecting an antigenic peptide:
  • 1. Although longer sequences can be used, many prefer a peptide of lengths 14-18.
  • 2. Sequences at the N- or C-terminus are frequently chosen.
  • 3. Hydrophilic sequences are preferred because they are readily soluble in an aqueous buffer.
  • 4. Hydorphilic sequences are likely on protein surface and are accessible.
  • 5. A cysteine is frequently added at the N- or C-terminus for conjugation, however, if there are two cysteines present, disulphide bonds may form inter- and intra-peptide leading to insolubility and structural alteration.
  • 6. Post-translation modification site should in general be avoided.
  • 7. Other suggestions include avoiding typtophan and selecting proline and tyrosine.
  • 8. Peptide should be checked for the possibility of raising a cross-reactive antibody.

    An analysis of hundreds of antigenic peptide sequences shows that although many researchers have respected these rules in selecting a peptide for raising an antibody, a few have found that exceptions also work. Abie does not emphasize on any of the particular rules, but gives a user a convenient interface in testing the effects of these rules. With Abie, users will find it is easier to decide which peptide is the best candidate for the protein of interest.

    A few tips

    Frequently, users will find that Abie gives too many potential candidates. Frequently, the reason is that there is a long stretch of amino acids that satisfies the criteria a user chooses. Fewer choices will be found if one restricts to a particular region of interest and/or start with a longer peptide length. After a user has decided on fewer choices, one could try to see whether shorter peptide sequences would satisfy the same criteria.

    Statistics of antigenic peptides

    Figure 2. Length distribution of a collection of antigenic peptides.
    Figure 3. Amino acid frequencies in a collection of antigenic peptides. Terminal cystines are excluded in the statistics.
    Figure 4. Frequencies of post-translation modification consensus in a collection of antigenic peptides.
    Home Products Order Contact
  • Copyright 2002-2004 Chang Bioscience, Inc. All rights reserved.