Related Experiment Video
Updated: Jul 8, 2026
08:59
Defining Substrate Specificities for Lipase and Phospholipase Candidates
Published on: November 23, 2016
DeepAden: an explainable machine learning method for predicting the substrate specificity of nonribosomal peptide
Jiaquan Huang1, Liangjun Ge1, Yaxin Wu1
1Center for Biological Science and Technology, Advanced Institute of Natural Sciences, Beijing Normal University, Zhuhai, Guangdong 519087, People's Republic of China.
Nucleic Acids Research
|July 6, 2026
Summary
DeepAden, a deep learning model, accurately predicts amino acid substrates for nonribosomal peptide synthetases (NRPS). This aids in discovering new microbial natural products and drug leads.
Area of Science:
- Biochemistry and Molecular Biology
- Computational Biology and Bioinformatics
- Natural Product Drug Discovery
Background:
- Microbial nonribosomal peptides (NRPs) are structurally diverse and a key source of drug leads.
- Nonribosomal peptide synthetases (NRPSs), specifically adenylation (A) domains, determine NRP core structures by activating amino acid substrates.
- Predicting A-domain substrate specificity is crucial for understanding NRP biosynthesis and discovering new compounds.
Purpose of the Study:
- To develop a deep learning model, DeepAden, for accurate prediction of A-domain substrate specificities.
- To enhance the understanding of NRP biosynthesis and facilitate the discovery of novel NRPs.
- To provide a practical computational tool for researchers in natural product drug discovery.
Main Methods:
- A two-stage deep learning approach was employed, utilizing a graph attention network (GAT) for binding pocket localization and representation.
- Pretrained language models and contrastive learning were used to encode pocket and substrate information.
- SHapley Additive exPlanations (SHAP)-guided data augmentation was applied to address class imbalance, especially for nonproteinogenic substrates.
Main Results:
- DeepAden demonstrated competitive performance against state-of-the-art methods on a benchmark dataset.
- The model successfully aided in annotating two Streptomyces NRPS gene clusters through substrate specificity predictions.
- DeepAden provides a robust method for A-domain substrate prediction and pocket localization.
Conclusions:
- DeepAden offers a practical and effective computational approach for predicting A-domain substrate specificities.
- This tool can accelerate the discovery and characterization of new NRPs with therapeutic potential.
- The DeepAden web server is publicly available at https://deepnp.site/ for broader research application.
More Related Videos
Related Concept Videos
01:56Riboswitches
Riboswitches are non-coding mRNA domains that regulate the transcription and translation of downstream genes without the help of proteins. Riboswitches bind directly to a metabolite and can form unique stem-loop or hairpin structures in response to the amount of the metabolite present. They have two distinct regions – a metabolite-binding aptamer and an expression platform.
The aptamer has high specificity for a particular metabolite which allows riboswitches to specifically regulate...
The aptamer has high specificity for a particular metabolite which allows riboswitches to specifically regulate...
01:13Induced-fit Model
Most chemical reactions in cells require enzymes—biological catalysts that speed up the reaction without being consumed or permanently changed. They reduce the activation energy needed to convert the reactants into products. Enzymes are proteins, that usually work by binding to a substrate—a reactant molecule that they act upon.
Enzymes exhibit substrate specificity, meaning that they can only bind to certain substrates. This is mainly determined by the shape and chemical characteristics of...
Enzymes exhibit substrate specificity, meaning that they can only bind to certain substrates. This is mainly determined by the shape and chemical characteristics of...
01:23Types of RNA
Overview
Three main types of RNA are involved in protein synthesis: messenger RNA (mRNA), transfer RNA (tRNA), and ribosomal RNA (rRNA). These RNAs perform diverse functions and can be broadly classified as protein-coding or non-coding RNA. Non-coding RNAs play important roles in the regulation of gene expression in response to developmental and environmental changes. Non-coding RNAs in prokaryotes can be manipulated to develop more effective antibacterial drugs for human or animal use.
RNA...
Three main types of RNA are involved in protein synthesis: messenger RNA (mRNA), transfer RNA (tRNA), and ribosomal RNA (rRNA). These RNAs perform diverse functions and can be broadly classified as protein-coding or non-coding RNA. Non-coding RNAs play important roles in the regulation of gene expression in response to developmental and environmental changes. Non-coding RNAs in prokaryotes can be manipulated to develop more effective antibacterial drugs for human or animal use.
RNA...
01:19Allosteric Proteins-ATCase
Binding sites linkages can regulate a protein's function. For example, enzyme activity is often regulated through a feedback mechanism where the end product of the biochemical process serves as an inhibitor.
Aspartate transcarbamoylase (ATCase) is a cytosolic enzyme that catalyzes the condensation of L-aspartate and carbamoyl phosphate to N-carbamoyl-L-aspartate. This reaction is the first step in pyrimidine biosynthesis. UTP and CTP, the end products of the pyrimidine synthesis pathway,...
Aspartate transcarbamoylase (ATCase) is a cytosolic enzyme that catalyzes the condensation of L-aspartate and carbamoyl phosphate to N-carbamoyl-L-aspartate. This reaction is the first step in pyrimidine biosynthesis. UTP and CTP, the end products of the pyrimidine synthesis pathway,...
02:26tRNA Activation
Aminoacyl-tRNA synthetases are present in both eukaryotes and bacteria. Though eukaryotes have 20 different aminoacyl-tRNA synthetases to couple to 20 amino acids, many bacteria do not have genes for all of these aminoacyl-tRNA synthetases. Despite this, they still use all 20 amino acids to synthesize their proteins. For instance, some bacteria do not have the gene encoding the enzyme that couples glutamine with its partner tRNA. In these organisms, one enzyme adds glutamic acid to all of the...
01:49Conserved Binding Sites
Many proteins’ biological role depends on their interactions with their ligands, small molecules that bind to specific locations on the protein known as ligand-binding sites. Ligand-binding sites are often conserved among homologous proteins as these sites are critical for protein function.
Binding sites are often located in large pockets, and if their location on a protein’s surface is unknown, it can be predicted using various approaches. The energetic method computationally analyses the...
Binding sites are often located in large pockets, and if their location on a protein’s surface is unknown, it can be predicted using various approaches. The energetic method computationally analyses the...