Welcome to AIPAMPDS
AIPAMPDS (AI-Powered Antimicrobial Peptide Discovery System) is an integrated tool for generating and screening
antimicrobial
peptides.
This platform combines state-of-the-art machine learning models to help researchers discover and evaluate novel
AMPs.
Features
🧬 AMP Generation
Generate novel antimicrobial peptides (AMPs) from scratch using GPT-based models
with three generation objectives:
- Broad-spectrum AMPs: peptides with activity against a wide range of bacterial
species.
- Gram-negative-specific AMPs: peptides optimized for activity against Gram-negative
bacteria.
- Gram-positive-specific AMPs: peptides optimized for activity against Gram-positive
bacteria.
Customize peptide length (10-50 amino acids) and quantity
(1-1000 sequences).
Generate AMPs
🔬 Comprehensive Screening
Screen your peptide sequences through multiple prediction models:
- AMP Identifier: Predict antimicrobial activity
- Activity Predictor: Evaluate antimicrobial activity against E.coli and
S.aureus
- Hemolysis Predictor: Evaluate hemolytic toxicity
Support for FASTA file upload or direct sequence input (max
10000 sequences).
Screen Sequences
⏱️ Job Tracking
Monitor the progress of your analysis jobs in real-time:
- View status of submitted tasks
- Track progress percentage
- Download results when complete
- Automatic cleanup after 30 days
Results are stored in CSV format with detailed predictions and
properties.
Check Job Status
How to Use
For AMP Generation:
- Navigate to the Generation page
- Select a target (Broad-spectrum, Gram-negative-specific, or Gram-positive-specific)
- Specify the desired peptide length (10-50 amino acids)
- Choose the number of sequences to generate (1-1000)
- Submit the job and note the Job ID
- Wait for the job to finish and view the results, or record the Job ID and check the results later on the
Job Status page
- Download the results CSV file when complete
For Sequence Screening:
- Navigate to the Screening page
- Either paste sequences directly or upload a FASTA file
- Submit the job and note the Job ID
- Wait for the job to finish and view the results, or record the Job ID and check the results later on the
Job Status page
- Download the results CSV file when complete
About the Models
Our platform integrates four specialized deep learning models:
- GPT-based Generator: Conditional language model for de novo AMP design
- PeptideBERT-LSTM: Binary classifier for AMP identification
- ETFC Model: Dual predictor for antimicrobial activity against bacteria
- LSTM-Attention: Hemolytic activity predictor for safety assessment
Important: These predictions are computational estimates.
High-scoring peptides should be validated experimentally before clinical or commercial use.
Note: Job results are automatically deleted after 30 days. Please download your results promptly.