We recently made large changes to the webserver infrastructure, so you might experience errors. Please report issues to health-master@dtu.dk

AnOxPePred - 1.0

Prediction of antioxidative properties of peptides using convolutional neural network

AnOxPePred predicts the antioxidative (free radical scavenging and ion chelating) properties of peptides using convolutional neural network.

Submission

Sequence submission: paste the sequence(s) or upload a local file

Instructions

The server requires protein or peptide sequence(s), depending on the selected “Analyse Mode”, in FASTA format and cannot handle nucleic acid sequences. The server is also limited to the 20 standard amino acids.

Guideline:

Paste sequence(s) in FASTA format into the field or upload a FASTA file via the “Choose file” button.
Choose the minimum and maximum length of peptides seen in output (this will only affect the “Protein(s) w/ Exhaustive Mode” or “Protein(s) w/ Proteolysis Mode” Analyse Mode’s). Default is 10-20 residues, and the limits of the algorithm are 2-30 residues.
Depending on the FASTA input and intention, choose one of three Analyse Mode’s; “Peptide Mode”, “Protein(s) w/ Exhaustive Mode” or “Protein(s) w/ Proteolysis Mode” (See Mode descriptions).
Click the submit button to start the prediction. Depending on the sample size and chosen Analyse Mode, computational time can range from a few seconds (Peptide Mode with few samples) to several hours (Protein(s) w/ Exhaustive Mode with many samples).
After the server successfully finishes the job, a Server Output page shows up. If an error happens during prediction a log will appear specifying the error. The default page is the Server Output, where a short description and summary of the prediction can be seen. For large inputs, the full results can be seen through the links.

Mode descriptions:

Peptide Mode

Predicts the activity directly from the peptides provide (no cutting up the sequences as with the other modes). The settings of Minimum and Maximum peptide length do not affect which peptides get predicted, but a standard limit of 2-30 residues is used for every peptide.

Protein(s) w/ Exhaustive Mode

Each input protein is cut into every possible peptide with the given lengths (the settings of Minimum and Maximum peptide length). This exhaustive mode thereby gives a comprehensive prediction of all peptides embedded in the proteins. To make the results easier to interpret, it is noted which proteins each sequence is present in and output files also include clustering of similar sequences (70% sequence identity).

In the case of many, large proteins, this mode might give to overwhelming results and the Protein(s) w/ Proteolysis Mode is in these cases recommended.

Protein(s) w/ Proteolysis Mode

The proteins are cleaved by combination of selected proteases (at least one must be selected). Resulting peptides outside given lengths (the settings of Minimum and Maximum peptide length) are deselected from the output. This mode considers that all cleavage sites are not always cleaved. All combinations of peptides being cleaved or not, are therefore included, to give the best overall results.

The cleavage specificities of each protease adhere to the following defined guidelines: Expasy Guidelines.

Abstract

AnOxPePred: Using deep learning for the prediction of antioxidative properties of peptides. (In preparation), (2019).
Olsen T.H., Yesiltas B., Marin F.I., Pertseva M., García-Moreno P.J., Gregersen S., Overgaard M.T., Jacobsen C., Lund O., Hansen E.B. and Marcatili P.
Nature Scientific Reports (December 2020). doi: 10.1038/s41598-020-78319-w

Abstract

Dietary antioxidants are an important preservative in food and have been suggested to help in disease prevention. With consumer demands for less synthetic and safer additives in food products, the food industry is searching for antioxidants that can be marketed as natural. Peptides derived from natural proteins show promise, as they are generally regarded as safe and potentially contain other beneficial bioactivities. Antioxidative peptides are usually obtained by testing various peptides derived from hydrolysis of proteins by a selection of proteases. This slow and cumbersome trial-and-error approach to identify antioxidative peptides has increased interest in developing computational approaches for prediction of antioxidant activity and thereby reduce laboratory work. A few antioxidant predictors exist, however, no tool predicting the antioxidative properties of peptides is, to the best of our knowledge, currently available as a web-server. We here present the AnOxPePred tool and web-server (http://services.bioinformatics.dtu.dk/service.php?AnOxPePred-1.0) that uses deep learning to predict the antioxidant properties of peptides. Our model was trained on a curated dataset consisting of experimentally-tested antioxidant and non-antioxidant peptides. For a variety of metrics our method displays a prediction performance better than a k-NN sequence identity-based approach. Furthermore, the developed tool will be a good benchmark for future predictors of antioxidant peptides.

Data Set

The data set utilized when training the convolutional neural networks can be downloaded here: 01_AO_dataset.csv

GETTING HELP

If you need help regarding technical issues (e.g. errors or missing results) contact Technical Support. Please include the name of the service and version (e.g. NetPhos-4.0). If the error occurs after the job has started running, please include the JOB ID (the long code that you see while the job is running).

If you have scientific questions (e.g. how the method works or how to interpret results), contact Correspondence.

Correspondence: Technical Support: