Peptide Resources

MLACP 2.0 is an upgraded machine learning tool for anticancer peptide (ACPs) prediction, constructing the first large non-redundant training and independent datasets, integrating diverse feature encodings (e.g., sequence physicochemical properties, structural features) and combining prediction results from seven conventional classifiers (including logistic regression, random forest, etc.), finally using convolutional neural network (CNN) for ensemble learning.

MLBP is a multi-label deep learning tool for predicting multi-functionalities of bioactive peptides, taking peptide sequence vectors as input to learn continuous feature representations via an embedding layer, and integrating convolutional neural network (CNN) with bidirectional gated recurrent unit (Bi-GRU) for sequence feature extraction. It simultaneously predicts anti-cancer, anti-diabetic, anti-hypertensive, anti-inflammatory, and anti-microbial activities.

MLCPP 2.0 is a predictor for cell-penetrating peptides and their uptake efficiency, integrating 17 sequence feature encoding algorithms and 7 machine learning classifiers to build a stacking model, with updated dataset and optimized features.

NeuroPedia is an encyclopedic database for neuropeptides, integrating peptide sequences, genomic locations, taxonomic information, and tandem mass spectrometry (MS/MS) spectral libraries of homologous neuropeptides across species. Addressing the challenges of neuropeptide identification due to atypical lengths (shorter/longer than tryptic peptides) and lack of ionization sites, the database enables searching MS/MS data against known neuropeptide sequences to improve identification sensitivity. Its spectral libraries support tandem with spectral matching tools, enhancing confidence through visual comparison of new and existing neuropeptide MS/MS spectra, providing mass spectrometry data support for neuroendocrine physiological and pathological research.

NeuroPep is a comprehensive resource for neuropeptides, housing 5,949 non-redundant entries from 493 species across 65 neuropeptide families (3,455 invertebrates, 2,406 vertebrates). Each entry provides source organism, tissue specificity, family classification, post-translational modifications, 3D structures (where available), and literature references, with physicochemical properties (amino acid composition, isoelectric point, molecular weight). Supporting keyword quick search and logical operator advanced search, it integrates sequence browsing, alignment, and mapping tools, facilitating neuropeptide physiological research and nervous system disease therapeutic target development.