Biology subjects

Lim, T. K. Y.

Publications and source records attributed to Lim, T. K. Y..

4 recordsLinked to original sources

bioRxiv2025

DeorphaNN: Virtual screening of GPCR peptide agonists using AlphaFold-predicted active state complexes and deep learning embeddings

Peptide-activated G protein-coupled receptors (GPCRs) regulate critical physiological processes such as metabolism, neural signalling, and endocrine function through their interaction with neuropeptides and peptide hormones. Despite their importance, identifying endogenous peptide agonists for GPCRs remains challenging, particularly for orphan receptors without known ligands. Recent advances in deep learning-based protein structure prediction, exemplified by AlphaFold (AF), have shown application beyond structural modelling, including protein-protein interaction prediction. Given that GPCR-peptide agonist interactions represent a specialized form of protein-protein interaction, we leveraged a dataset of experimentally validated agonist and non-agonist GPCR-peptide interactions from Caenorhabditis elegans to evaluate AF-Multimers ability to distinguish agonist-bound complexes. When modelling GPCR-peptide complexes, AF-Multimer confidence metrics partially discriminate agonist from non-agonist interactions, with improved discrimination achieved by utilizing AF-Multistate-derived active-state templates. We further investigated whether embeddings from the hidden layer of AF-Multimers neural network could distinguish agonist from non-agonist complexes. Feature performance analysis reveals that AF- Multimers pair representations outperform single representations, with distinct subregions of the pair representation providing complementary predictive signals. Building on these insights, we developed DeorphaNN, a graph neural network integrating active-state GPCR-peptide structural predictions, interatomic interactions, and deep learning embeddings to prioritize GPCR-peptide agonist interactions. DeorphaNN generalizes across datasets derived from different species, including annelids and humans, and successfully uncovered peptide agonists for two orphan GPCRs. DeorphaNN offers a novel computational resource to accelerate deorphanization by prioritizing GPCR-peptide agonist candidates for AI-guided experimental validation.

bioinformatics↗

bioRxiv2024

Non-immunogenic and reversible transgene expression using self-amplifying RNA

Self-amplifying RNA (saRNA) has the potential to provide durable, non-integrating transgene expression for transient gene therapy. However, its auto-replicative nature mimics viral infection, triggering innate immune responses that shut down cap-dependent translation, degrade cellular mRNA, induce cell death, and release cytokines. In non-immunotherapy applications, this immune activation is undesirable as it limits transgene expression, induces unintended changes in host gene expression, depletes transfected cells, and promotes inflammation--ultimately undermining therapeutic outcomes. Moreover, the use of exogenous immune suppressants to mitigate these effects often increases treatment complexity and the risk of unintended systemic side effects. To address these challenges, we developed a strategy to encode broad-spectrum innate immune suppression directly within saRNA. This approach leverages cap-independent translation to bypass saRNA-triggered translation shutdown, enabling the expression of multiple inhibitors targeting diverse double-stranded RNA-sensing and inflammatory signalling pathways. In mouse primary fibroblast-like synoviocytes--key cells in joint pathologies--this strategy eliminates the need for external immune inhibitors, reduces cytotoxicity and antiviral cytokine secretion, and enables sustained transgene expression that can be controlled with a small-molecule antiviral. Together, these findings support the development of reversible immune-evasive saRNA constructs for transient gene therapy applications that avoid persistent immune activation and eliminate the need for external immune suppressants. GRAPHICAL ABSTRACT O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=175 SRC="FIGDIR/small/614636v6_ufig1.gif" ALT="Figure 1"> View larger version (35K): org.highwire.dtl.DTLVardef@18c1f36org.highwire.dtl.DTLVardef@554f8borg.highwire.dtl.DTLVardef@d9370dorg.highwire.dtl.DTLVardef@ad5ff2_HPS_FORMAT_FIGEXP M_FIG C_FIG

bioengineering↗

bioRxiv2022

CysPresso: Prediction of cysteine-dense peptide expression in mammalian cells using deep learning protein representations

Background: Cysteine-dense peptides (CDPs) are an attractive pharmaceutical scaffold that display extreme biochemical properties, low immunogenicity, and the ability to bind targets with high affinity and selectivity. While many CDPs have potential and confirmed therapeutic uses, synthesis of CDPs is a challenge. Recent advances have made the recombinant expression of CDPs a viable alternative to chemical synthesis. Moreover, identifying CDPs that can be expressed in mammalian cells is crucial in predicting their compatibility with gene therapy and mRNA therapy. Currently, we lack the ability to identify CDPs that will express recombinantly in mammalian cells without labour intensive experimentation. To address this, we developed CysPresso, a novel machine learning model that predicts recombinant expression of CDPs based on primary sequence. Results: We tested various protein representations generated by deep learning algorithms (SeqVec, proteInfer, AlphaFold2) for their suitability in predicting CDP expression and found that AlphaFold2 representations possessed the best predictive features. We then optimized the model by concatenation of AlphaFold2 representations, time series transformation with random convolutional kernels, and dataset partitioning. Conclusion: Our novel model, CysPresso, is the first to successfully predict recombinant CDP expression in mammalian cells and is particularly well suited for predicting recombinant expression of knottin peptides. When preprocessing the deep learning protein representation for supervised machine learning, we found that random convolutional kernel transformation preserves more pertinent information relevant for predicting expressibility than embedding averaging. Our study showcases the applicability of deep learning-based protein representations, such as those provided by AlphaFold2, in tasks beyond structure prediction.

bioinformatics↗

bioRxiv2020

In vivo imaging of microglia-mediated axonal pruning and modulation by the complement system

Partial phagocytosis--called trogocytosis--of axons by microglia has been documented in ex vivo preparations but has not been directly observed in vivo. The mechanisms that modulate microglial trogocytosis of axons and its function in neural circuit development remain poorly understood. Here we directly observe axon trogocytosis by microglia in vivo in the developing Xenopus laevis retinotectal circuit. We show that microglia regulate pruning of retinal ganglion cell axons and are important for proper behavioral response to dark and bright looming stimuli. We identify amphibian regulator of complement activation 3, a homolog of human CD46, as a neuronally-expressed synapse-associated complement inhibitory molecule that inhibits trogocytosis and axonal pruning. Using a membrane-bound complement C3 fusion protein, we demonstrate that enhancing complement activity enhances axonal pruning. Our results support the model that microglia remodel axons via trogocytosis and that neurons can control this process through expression of complement inhibitory proteins.

neuroscience↗