Datasets are provided here for the purpose of reproducibility and future method testing. If you are looking for information about AMP sequences, you are recommended to visit other AMP databases such as ADP, CAMP, LAMP, etc. Our datasets here were collected from these databases and filtered out sequences with non-natural amino acids.

Anti-Microbial Peptides

Data used for constructing the AmPEP prediction model.

Unique data collected from ADP3, CAMPR3, LAMP. All non-natural amino acids were removed.
Generated from Uniprot sequences without annotation of AMP, membrane, toxic, secretory, defensing, antibiotic, anticancer, antiviral and antifungal.

Benchmark datasets from Xiao et al. (iAMP-2L) for methods comparison can be downloaded from here.


AmPEP: Sequence-based prediction of antimicrobial peptides using distribution patterns of amino acid properties and random forest.

Scientific Reports, 1697 (2018).

Pratiti Bhadra, Jielu Yan, Jinyan Li, Simon Fong, and Shirley W. I. Siu.*

Short Anti-Microbial Peptides

Data is filtered from our AmPEP dataset, include sequences only with 5-30 AA in length. This dataset is used for constructing the Deep-AmPEP30 and RF-AmPEP30 prediction models. An independent dataset was constructed as benchmark to compare model performances with other existing methods.

Train Dataset
1529 positives and 1529 negatives
Test Dataset (Benchmark)
94 positives and 94 negatives

Deep-AmPEP30: Improve short antimicrobial peptides prediction with deep learning.

Accepted for publication in Molecular Therapy – Nucleic Acids.

Jielu Yan, Pratiti Bhadra, Ang Li, Pooja Sethiya, Longguang Qin, Hio Kuan Tai, Koon Ho Wong, and Shirley W. I. Siu*