Protein post-translational modifications (PTMs) are critical for regulating protein function and are closely linked to disease mechanisms. In-depth research and precise prediction of PTMs are vital for understanding life mechanisms, screening disease biomarkers, and identifying drug targets. Artificial intelligence (AI) approaches for PTM site prediction offer complementary advantages to traditional experimental methods, providing high-throughput and cost-effective screening that can prioritize candidate sites for further validation. This paper reviews advances in PTM site prediction since 2012, focusing on machine learning and deep learning techniques. It analyzes more than 500 relevant studies and categorizes 36 types of PTMs. Additionally, the paper briefly outlines core contents such as database resources related to PTMs, commonly used feature extraction methods, and major classification algorithms. In addition, 36 representative recent studies on PTMs have been carefully selected for in-depth analysis. The findings indicate that current machine learning-based PTM research employs multivariate feature extraction and construct composite models to enhance prediction performance. Finally, keyword visualization using CiteSpace identifies emerging research hotspots and future directions for PTM site prediction.

Designing protein sequences with desired properties is a fundamental task in protein engineering. Recent advances in deep generative models have greatly accelerated this design process. However, most existing models face the issue of distribution centralization and focus on local compositional statistics of natural sequences instead of the global semantic organization of protein space, which confines their generation to specific regions of the distribution. These problems are amplified for functional proteins, whose sequence patterns strongly correlate with semantic representations and exhibit a long-tailed functional distribution, causing existing models to miss semantic regions associated with rare but essential functions. Here, we propose ProtFlow, a generative model designed for comprehensive semantic distribution learning of protein sequences, enabling high-quality sequence generation. ProtFlow employs a rectified flow matching algorithm to efficiently capture the underlying semantic distribution of the protein design manifold, and introduces a reflow technique enabling one-step sequence generation. We construct a semantic integration network to reorganize the representation space of large protein language models, facilitating stable and compact incorporation of global protein semantics. We pretrain ProtFlow on 2.6M peptide sequences and fine-tune it on antimicrobial peptides (AMPs), a representative class of therapeutic proteins exhibiting unevenly distributed activities across pathogen targets. Experiments show that ProtFlow outperforms state-of-the-art methods in generating high-quality peptides, and AMPs with desirable activity profiles across a range of pathogens, particularly against underrepresented bacterial species. These results demonstrate ProtFlow’s effectiveness in capturing the full training distribution and its potential as a general framework for computational protein design.