How many natural antimicrobial peptides have been identified with known activity?
When we talk about "how many AMPs have been discovered", we mean natural peptides because the number of synthetic and predicted peptides are unlimited. According to the APD6, there are 3379 natural AMPs with known antimicrobial activity as of Dec 2025. This list, started in 2003, has been continuously updated and refined for over 20 years. The refined peptide lists for downloads do not include over 200 natural peptides without antimicrobial activity. They can be downloaded here.
With the expansion of this database, how do I search my favorite APD3 with a unique set of natural AMPs?
The APD3 focused on natural antimicrobial peptides. Consdiering the significnce of natural AMPs, this database feature is retained in the APD6. To get an updated list, you can search "natural AMPs" in the NAME field of the current database. If you keep "natural AMPs" in the name field, all your search results will be related to such peptides only.
How can I search machine learning (ML)/Artificial intelligence (AI) predicted peptides?
For a subgroup of antimicrobial peptides predicted by machine learning (shallow learning) or AI (deep learning) and validated experimentally with antimicrobial activity, you can search such peptides in the Source field in the the APD6 using "AI predicted". As of Dec 31, 2025, we found 281 such peptides.
How can I obtain AMPs with MIC values for AI prediction?
Multiple researchers have asked the same question. We are unifying the data format at the moment, validating the data extraction programs, and will inform you when ready.
How can I search antimicrobial peptides discovered from microbiota?
The APD first included this info in the field of Source organism (Protein Sci 2022). You can search "microbiota" here by entering microbiota:gut or microbiota:skin.
How can I search cyclic AMPs in the APD?
Antimicrobial peptides can be cyclic via different chemical bonds, which form the basis for our proposed universal peptide classification method initially published in 2015. In the APD, cyclic peptides are split into three major classes: (1) sidechain-to-sidechain connected (search using UCSS, 1631 entries in Dec 2024); (2) sidechain-backbone connected (search using UCSB, 64 peptides); and (3) head-to-tail connected (search using UCBB, 231 peptides). Please enter your search terms (e.g., UCBB) into the NAME field of the APD search interface. If you want to focus on natural antimicrobial peptides, please also enter "Natural AMPs" into another box of the Name field here.
How can I do a calculation of the peptide properties such as GRAVY and molecule weight when the sequence is already in the APD?
For properties independent of peptide sequence, you can do so by shuffling the position of one amino acid followed by calculation here.
Is it possible for us to calculate and predict a peptide sequence with more than 50 amino acid residues?
Yes, you can do it here.
Can we search antimicrobial peptides from bacteria, plants, or animals only?
Yes, you can do this by entering "plants" into one of the search boxes in the NAME field here. Then you can search the info in the usual way. Likewise, you can search for a list of AMPs from bacteria (i.e. bacteriocins) or a vareity of animal classes (e.g., mammals, insects, spiders, and scorpions). See the APD face page for a more complete list of these animal groups.
How do I obtain a list of AMPs that bind to lipopolysaccharides (LPS) or endotoxin?
You can search such peptides by entering the code "BBL" into one of the search boxes in the NAME field here. For additional codes for AMP-binding partners, go to the Glossary page.
Can I search AMPs in the form of dimer or two chains?
Yes, the APD FIRST inlcuded such information in the APD2. You can do so by entering "dimer" or "two-chain" into the NAME field followed by database search. Homodimer and Heterodimer peptides can be searched separately.
Can I search AMPs based on the year of publication?
Yes. This function currently shares the same search box with the author search here. A plot of AMPs discovered annually can be found online here (Wang, G. 2013 Pharmaceuticals 6: 728-758).
Does the database contain a link to the PubMed for each peptide entry?
Yes, this has been enabled for new entries and the links for the old entries are under construction. Additional references are included in the "Additional Info" field, which is under constant reformatting for clarity.
What is Boman Index?
We first named and programmed the Boman index in 2003 in the APD. It was originally called protein-binding potential by Hans Boman. The Boman index is the sum of free energies of the respective side chains for transfer from cyclohexane to water, taken from Radzeka and Wolfenden, and divided by the total number of the residues of an antimicrobial peptide. The calculated values are negative (except for the hybrid peptide), but the + and - are reversed [Boman, H.G. (2003) J. Inter. Med. 254:197-215].
How can I download all peptides or a defined class of AMPs in the FASTA format?
This download has been enabled upon users' request to promote research and education. For all the peptides, do an empty search in the APD followed by downloading the entire list by clicking on download button at bottom of the peptide list. Likewise, you can search any list of peptides based on source (life kindoms), structure, activity, targets, and so on. The 2024 natural AMP lists cover some peptide groups with over 100 entries.
How can I make use of the APD and its downloads properly?
You're welcome to use the APD downloads as long as you acknowledge the source of the information by citing the APD. If you have used the APD information in your article, it is best you indicate the source by providing the APD URL.
How can I submit my newly discovered peptides to the APD database?
We appreciate your input. The easiest way is to email Dr. Wang a copy of your manuscript/publihsed article (email: gwang@unmc.edu). In particular, we encourage you to include your peptide sequences in your email. This will minimize potential errors during data registration. The APD acknowledges your contributions by including "provided by YOU" in each entry. As of Dec 2025, APD users, fans, and friends have assisted with the registration of 316 AMPs in the APD.
The APD is very helpful to my research. How do I cite it?
You can cite the following articles that describe the APD:
Wang, G, Li, X. and Wang, Z.(2016) APD3: the antimicrobial peptide database as a tool for research and education. Nucleic Acids Research 44, D1087-D1093.
You can also cite the APD6 that reports a new information pipeline for AMPs.
Important Note: If your search results differ dramatically from the statistics on the main page of the APD, your web browser may need update.