Skip to main content
recep.adiyaman
Vision About Experience Education Skills Publications Tools Newsletter Updates Podcasts Blog Contact CV Subscribe
All Issues
Older Newer
Daily Signal May 11, 2026 · 10 min read

Issue #105: Efficient Adaptation of Structure-Aware Protein Language Models for Diverse Protein Applications.

Protein Design Digest #105: scDock: streamlining drug discovery targeting cell-cell communication vi…

Share X LinkedIn
Protein Design Daily

Building something in Protein Design?

I love collaborating on new challenges. Let's build together.

LinkedIn Email

Subscribe to Protein Design Digest

Daily curated signals from arXiv, PubMed, and BioRxiv.

Signal of the Day

Efficient Adaptation of Structure-Aware Protein Language Models for Diverse Protein Applications.

Protein language models (PLMs) trained solely on sequence data have significantly advanced our understanding of protein biology and achieved remarkable performance in protein prediction tasks. However, their lack of three-dimensional (3D) structural features limits their predictive power in applications that rely heavily on 3D conformation. To address this limitation, we developed two structure-aware PLMs, S-PLM1 and S-PLM2, that employ multi-view contrastive learning to align protein sequences with their 3D structures in a unified latent space. S-PLM1 represents structural information using contact maps encoded by a pretrained Swin-Transformer, while S-PLM2 directly encodes 3D backbone coordinates through a Geometric Vector Perceptron (GVP)-based model. The paired sequence-structure data were obtained from AlphaFoldDB. For both models, we designed efficient tuning strategies that enable optimal performance with minimal computational cost. Here, we present detailed protocols for adapting S-PLM1 and S-PLM2 for diverse protein applications. The protocols provide step-by-step guidance on generating structure-aware representations from S-PLMs, fine-tuning them for various protein prediction tasks, and using S-PLM2 to produce structure embeddings for structure-based downstream analyses. We also provide source code and Google Colab implementations for easy customization and deployment. © 2026 Wiley Periodicals LLC. Basic Protocol 1: Generating structure-aware representations of protein sequences Basic Protocol 2: Efficient tuning of structure-aware protein language models for diverse protein applications Basic Protocol 3: Using S-PLM2 to generate protein structure representations and conduct structure-based clustering Support Protocol: Google Colab quick start notebooks.

Why this matters:

Also Worth Reading

Simulated construction of tilmicosin nucleic acid aptamers based on molecular docking and molecular dynamics techniques.

Traditional aptamer screening methods often prove ineffective for small molecule targets, primarily due to the inherent structural limitations of such compounds. Their simple architecture, limited functional groups, and restricted spatial complexity drastically reduce the probability of identifying nucleic acid sequences that bind with both high affinity and specificity. Consequently, the screening process becomes inefficient and labor-intensive, frequently failing to yield aptamers of satisfactory performance for practical applications. This represents a significant technical hurdle in expanding the use of aptamers in small-molecule detection and therapeutics. Based on this, this study innovatively proposes an aptamer design method based on single-nucleotide docking assembly, using the small molecule temicloxacin as an example. Through molecular dynamics simulations (50 ns, RMSD convergence threshold of 0.15 nm), the dynamic conformational characteristics of tilmicosin were analyzed. Subsequently, saturated docking was performed on four classes of mononucleotides, screening out 32 high-affinity mononucleotides (atomic contact distance ≤4 Å). Methods such as depth-first search algorithm (DFS) and weighted graph theory model were introduced to obtain the representative single nucleotides of eight classes of functional modules and linkage assembly, and finally 63 non-redundant candidate sequences were screened. Molecular docking results indicate that the optimal aptamer Til-14 exhibits high binding affinity with tilmicosin. with an affinity of 298.16 ± 95.588 nM measured via SYBR Green I fluorescence assay. Colloidal gold colorimetric analysis confirmed its high affinity (Kd = 279.323 ± 87.234 nM) and excellent specificity. This innovative method successfully addresses the key limitations of the traditional SELEX process in screening aptamers for small molecule targets. By enhancing the efficiency and specificity of selection, it not only facilitates the discovery of high-performance aptamers but also establishes a novel, generalizable framework for the construction of nucleic acid aptamers targeting other small molecules.

Exploring the Molecular Mechanism of Shaoyao Gancao Decoction for Trigeminal Neuralgia Based on Network Pharmacology, Molecular Docking, and Molecular Dynamics Simulation.

Background: Shaoyao Gancao Decoction (SGD) has demonstrated a broad spectrum of analgesic effects and has found application in the management of trigeminal neuralgia (TN). Nonetheless, the underlying molecular mechanisms of this therapeutic intervention remain poorly understood. Objective : This study is designed to elucidate the molecular mechanism of SGD in treating TN by employing an integrated approach that combines network pharmacology, molecular docking, and molecular dynamics simulation. Methods: The active ingredients and associated targets within SGD were screened from the TCMSP database. TN-related targets were extracted from GeneCards, OMIM, CTD, and DisGeNET databases. Subsequently, we constructed a comprehensive TN-SGD-herbs-ingredients-targets network by employing Cytoscape software for visualization. A protein-protein interaction (PPI) network was constructed using the STRING database and further analyzed with Cytoscape, from which we identified pivotal hub genes using three distinct Cytoscape plugins. GO and KEGG enrichment analyses were carried out utilizing R software. Then, molecular docking was executed using AutoDock Vina, and docking results were visualized and augmented with molecular dynamics simulations utilizing BIOVIA Discovery Studio software. Finally, in vitro experiments verified the anti-inflammatory effect of SGD on LPS-treated BV2 cells. Results: A total of 103 active ingredients within SGD, 332 targets associated with TN, and 68 potential therapeutic targets were obtained. We constructed a TN-SGD-herbs-ingredients-targets network and obtained a PPI network of potential therapeutic targets. Then, we extracted seven hub genes from the potential therapeutic targets, including ESR1, JUN, TP53, STAT3, BCL2, AKT1, and ESR2. GO enrichment analyses indicated that SGD affected multiple biological processes and functions, such as responses to xenobiotic stimuli, membrane rafts, DNA binding, and transcription factor binding. KEGG pathway analyses revealed that lipid and atherosclerosis, the AGE-RAGE signaling pathway in diabetic complications, and chemical carcinogenesis-receptor activation were mainly involved in the therapeutic effects of SGD on TN. Importantly, molecular docking analysis demonstrated substantial binding affinities between the top eight ranked active ingredients and the seven identified hub genes. Furthermore, molecular dynamics simulations validated the binding activity between shinpterocarpin and ESR2. Finally, SGD decreased the levels of TNF-α, IL-1β, and IL-6 and regulated protein expression of ESR1 and ESR2 on LPS-treated BV2 cells, indicating that SGD exerted anti-inflammatory effect on microglia. Conclusion: This study offers valuable insights into the active ingredients of SGD and elucidates their potential molecular mechanisms in the treatment of TN. The findings presented herein lay the groundwork for the development of anti-TN agents rooted in the constituents of SGD.

[Exploring the therapeutic targets and molecular mechanisms of pimecrolimus in the treatment of oral lichen planus based on network pharmacology, machine learning, and molecular docking].

Objective: To systematically investigate the potential targets and therapeutic mechanisms of pimecrolimus in the treatment of oral lichen planus (OLP) using network pharmacology, machine learning, and molecular docking approaches. Methods: Pimecrolimus and OLP-related targets were retrieved from SwissTargetPrediction, Comparative Toxicogenomics Database, and GeneCards database. Intersecting genes underwent gene ontology and Kyoto Encyclopedia of Genes and Genomes pathway enrichment analyses. A protein-protein interaction (PPI) network was constructed, with eXtreme Gradient Boosting algorithm and SHapley Additive exPlanations employed to evaluate target importance. Molecular docking validated binding affinity between pimecrolimus and core targets. Results: Twenty intersecting targets were identified, primarily enriched in inflammatory pathways, including phosphatidylinositol 3-kinase (PI3K)-protein kinase B (Akt) signaling pathway, tumor necrosis factor (TNF) signaling pathway, and nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) signaling pathway. PPI and machine learning identified key targets including matrix metalloproteinase 2 (MMP2), mechanistic target of rapamycin (MTOR), and erb-b2 receptor tyrosine kinase 2 (ERBB2). The expression levels of MMP2 and MTOR in erosive OLP (6.67 and 3.54, respectively) were significantly higher than those in non-erosive OLP (4.39 and 3.27, respectively) ( P P =0.012). In contrast, the expression level of ERBB2 in non-erosive OLP (5.55) was significantly higher than that in erosive OLP (4.86) ( P =0.001). Molecular docking demonstrated that pimecrolimus had good binding activity with all the above targets, with binding energies all lower than -5 kcal/mol. Conclusions: This study systematically demonstrates that pimecrolimus may interfere with OLP progression by synergistically regulating key targets such as MMP2 (acting on erosive OLP), MTOR (acting on erosive OLP), and ERBB2 (acting on non-erosive OLP), and inhibiting the activity of inflammatory pathways including PI3K-Akt, TNF, and NF-κB.

Research & AI Updates

From the Industry

Quick Reads

Scaffold-based evaluation metrics for fair comparison of molecular generators.

Molecular generators enable the exploration of chemical space to identify novel compounds with desirable properties. Read more →

Exploring the antibacterial mechanism and application of lavender essential oil based on network pharmacology and molecular docking.

In this study, essential oil was extracted from lavender using steam distillation, and its chemical components were identified by gas chromatography-mass spectrometry (GC-MS). Read more →

Indicaxanthin Mitigates Hepatic Ischemia Reperfusion Injury in Rats via Bax/Bcl-2 Modulation: Experimental Evidence Supported by Molecular Docking and Dynamics Simulation.

Indicaxanthin, a prominent betalain pigment known for its bioactivity and safety, has long been valued in traditional medicine for its antioxidant and anti-inflammatory effects. Read more →

Efficient Adaptation of Structure-Aware Protein Language Models for Diverse Protein Applications.

Protein language models (PLMs) trained solely on sequence data have significantly advanced our understanding of protein biology and achieved remarkable performance in protein prediction tasks. Read more →

Design, synthesis, and molecular docking evaluation of novel pyrimidine-thiadiazole glycosides as potential anticancer agents.

A series of novel pyrimidine-thiadiazole conjugates incorporating carbohydrate-based 1,3,4-thiadiazole thioglycosides were synthesized and analyzed using (IR), (1H-NMR) and (13C-NMR) spectroscopy. Read more →

Synthesis and Characterization of Mannich-Derived 1,4-Naphthoquinones From Lawsone: Photophysical Properties, Molecular Docking, BSA/DNA Interactions, and Antioxidant Activity.

The present study focuses on the synthesis of eight novel 2-hydroxy-1,4-naphthoquinone (lawsone) derivatives in moderate to high yields (31%-90%) via the Mannich reaction. Read more →

Multispectral, Molecular Docking, and Dynamics Simulation Studies of Secalonic Acid F Binding to Human Serum Albumin.

Secalonic acid F (SAF) is a fungal secondary metabolite with broad pharmacological activities. Read more →

First GC/MS Profiling and Cytotoxic Evaluation of Physalis acutifolia (Solanaceae) Essential Oils: Insights From Molecular Docking and ADME Predictions.

The Solanaceae family, particularly the genus Physalis, is a rich source of anticancer bioactive compounds. Read more →

Pipeline Tip

Use Snakemake for reproducible end-to-end protein design workflows.

Resources & Tools

Deep learning is not a magic wand, but a powerful lens for structural biology. — Recep Adiyaman

BS HF DK