ACS Bio & Med Chem Au最新文献

{"title":"Leveraging the Capsular Polysaccharide Synthesis Pathway in <i>Streptococcus pneumoniae</i> as a Genetic Glycoengineering Platform.","authors":"Cheryl Kang-Rou Wong, Ye-Yu Chun, Tong Su, Lok-To Sham","doi":"10.1021/acsbiomedchemau.5c00010","DOIUrl":"10.1021/acsbiomedchemau.5c00010","url":null,"abstract":"<p><p>Engineering carbohydrates in living cells is one of the overarching goals of biology. In this Perspective, we discuss recent work in response to this challenge. Compared with eukaryotic cells, bacteria are fast-growing and genetically tractable. At the species level, glycans in prokaryotes are highly variable, contrasting with the homogeneity of surface glycans, such as capsular polysaccharides (CPSs), at the strain level. We exploited the conditional essentiality of the CPS synthesis pathway in <i>Streptococcus pneumoniae</i> to overcome the challenges of biochemically monitoring the engineered glycan products. While this strategy seems feasible, this glycoengineering platform is limited by the specificity of the capsule transporters and the glycosyltransferase inventories that can be introduced into the pneumococcus. Mutants that relax transporter specificity have been isolated, enabling us to inactivate otherwise essential glycosyltransferases. Ongoing work aims to harness this technology to synthesize medically relevant glycans, including Lewis antigens and tumor markers.</p>","PeriodicalId":29802,"journal":{"name":"ACS Bio & Med Chem Au","volume":"5 3","pages":"342-349"},"PeriodicalIF":3.8,"publicationDate":"2025-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12186847/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144486243","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Leveraging the Capsular Polysaccharide Synthesis Pathway in Streptococcus pneumoniae as a Genetic Glycoengineering Platform","authors":"Cheryl Kang-Rou Wong,&nbsp;Ye-Yu Chun,&nbsp;Tong Su and Lok-To Sham*,&nbsp;","doi":"10.1021/acsbiomedchemau.5c0001010.1021/acsbiomedchemau.5c00010","DOIUrl":"https://doi.org/10.1021/acsbiomedchemau.5c00010https://doi.org/10.1021/acsbiomedchemau.5c00010","url":null,"abstract":"<p >Engineering carbohydrates in living cells is one of the overarching goals of biology. In this Perspective, we discuss recent work in response to this challenge. Compared with eukaryotic cells, bacteria are fast-growing and genetically tractable. At the species level, glycans in prokaryotes are highly variable, contrasting with the homogeneity of surface glycans, such as capsular polysaccharides (CPSs), at the strain level. We exploited the conditional essentiality of the CPS synthesis pathway in <i>Streptococcus pneumoniae</i> to overcome the challenges of biochemically monitoring the engineered glycan products. While this strategy seems feasible, this glycoengineering platform is limited by the specificity of the capsule transporters and the glycosyltransferase inventories that can be introduced into the pneumococcus. Mutants that relax transporter specificity have been isolated, enabling us to inactivate otherwise essential glycosyltransferases. Ongoing work aims to harness this technology to synthesize medically relevant glycans, including Lewis antigens and tumor markers.</p>","PeriodicalId":29802,"journal":{"name":"ACS Bio & Med Chem Au","volume":"5 3","pages":"342–349 342–349"},"PeriodicalIF":3.8,"publicationDate":"2025-05-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://pubs.acs.org/doi/epdf/10.1021/acsbiomedchemau.5c00010","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144305832","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Unraveling the GM1 Specificity of Galectin-1 Binding to Lipid Membranes","authors":"Federica Scollo*,&nbsp;Waldemar Kulig,&nbsp;Gabriele Nicita,&nbsp;Anna-Kristin Ludwig,&nbsp;Joana C. Ricardo,&nbsp;Valeria Zito,&nbsp;Peter Kapusta,&nbsp;Ilpo Vattulainen,&nbsp;Marek Cebecauer,&nbsp;Hans-Joachim Gabius,&nbsp;Herbert Kaltner,&nbsp;Giuseppe Maccarrone* and Martin Hof*,&nbsp;","doi":"10.1021/acsbiomedchemau.5c0004010.1021/acsbiomedchemau.5c00040","DOIUrl":"https://doi.org/10.1021/acsbiomedchemau.5c00040https://doi.org/10.1021/acsbiomedchemau.5c00040","url":null,"abstract":"<p >Galectin-1 (Gal-1) is a galactose-binding protein involved in various cellular functions. Gal-1’s activity has been suggested to be connected to two molecular concepts, which are, however, lacking experimental proof: a) enhanced binding affinity of Gal-1 toward membranes containing monosialotetrahexosylganglioside (GM₁) over disialoganglioside GD₁a and b) cross-linking of GM₁’s by homodimers of Gal-1. We provide evidence about the specificity and the nature of the interaction of Gal-1 with model membranes containing GM₁ or GD₁a, employing a broad panel of fluorescence-based and label-free experimental techniques, complemented by atomistic biomolecular simulations. Our study demonstrates that Gal-1 indeed binds specifically to GM₁ and not to GD₁a when embedded in membranes over a wide range of concentrations (i.e., 30 nM to 20 μM). The apparent binding constant is about tens of micromoles. On the other hand, no evidence of Gal-1/GM₁ cross-linking was observed. Our findings suggest that cross-linking does not result from sole interactions between GM₁ and Gal-1, indicating that in a physiological context, additional triggers are needed, which shift the GM₁/Gal-1 equilibria toward the membrane-bound homodimeric Gal-1.</p>","PeriodicalId":29802,"journal":{"name":"ACS Bio & Med Chem Au","volume":"5 3","pages":"415–426 415–426"},"PeriodicalIF":3.8,"publicationDate":"2025-05-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://pubs.acs.org/doi/epdf/10.1021/acsbiomedchemau.5c00040","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144305831","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Unraveling the GM₁ Specificity of Galectin‑1 Binding to Lipid Membranes.","authors":"Federica Scollo, Waldemar Kulig, Gabriele Nicita, Anna-Kristin Ludwig, Joana C Ricardo, Valeria Zito, Peter Kapusta, Ilpo Vattulainen, Marek Cebecauer, Hans-Joachim Gabius, Herbert Kaltner, Giuseppe Maccarrone, Martin Hof","doi":"10.1021/acsbiomedchemau.5c00040","DOIUrl":"10.1021/acsbiomedchemau.5c00040","url":null,"abstract":"<p><p>Galectin-1 (Gal-1) is a galactose-binding protein involved in various cellular functions. Gal-1's activity has been suggested to be connected to two molecular concepts, which are, however, lacking experimental proof: a) enhanced binding affinity of Gal-1 toward membranes containing monosialotetrahexosylganglioside (GM₁) over disialoganglioside GD₁a and b) cross-linking of GM₁'s by homodimers of Gal-1. We provide evidence about the specificity and the nature of the interaction of Gal-1 with model membranes containing GM₁ or GD₁a, employing a broad panel of fluorescence-based and label-free experimental techniques, complemented by atomistic biomolecular simulations. Our study demonstrates that Gal-1 indeed binds specifically to GM₁ and not to GD₁a when embedded in membranes over a wide range of concentrations (i.e., 30 nM to 20 μM). The apparent binding constant is about tens of micromoles. On the other hand, no evidence of Gal-1/GM₁ cross-linking was observed. Our findings suggest that cross-linking does not result from sole interactions between GM₁ and Gal-1, indicating that in a physiological context, additional triggers are needed, which shift the GM₁/Gal-1 equilibria toward the membrane-bound homodimeric Gal-1.</p>","PeriodicalId":29802,"journal":{"name":"ACS Bio & Med Chem Au","volume":"5 3","pages":"415-426"},"PeriodicalIF":3.8,"publicationDate":"2025-05-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12183518/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144486250","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Radical Fluoromethylation Enabled by Cobalamin-Dependent Radical SAM Enzymes","authors":"Syam Sundar Neti,&nbsp;Bo Wang*,&nbsp;Jiayuan Cui,&nbsp;David F. Iwig,&nbsp;Nicholas J. York,&nbsp;Anthony J. Blaszczyk,&nbsp;Matthew R. Bauerle and Squire J. Booker*,&nbsp;","doi":"10.1021/acsbiomedchemau.5c0006210.1021/acsbiomedchemau.5c00062","DOIUrl":"https://doi.org/10.1021/acsbiomedchemau.5c00062https://doi.org/10.1021/acsbiomedchemau.5c00062","url":null,"abstract":"<p >Fluorine is an important atom in many drugs because it can improve the efficacy and metabolic stability of many molecules. Strategies to incorporate monofluoromethyl groups in drugs have been limited and have received less attention than strategies for difluoromethylation or trifluoromethylation. Previously, we and others reported the enzymatic monofluoromethylation of several biologically relevant metabolites based on the transfer of a fluoromethyl group from analogs of <i>S</i>-adenosylmethionine (SAM) to various nucleophiles (carbon, oxygen, nitrogen, sulfur, and carbon) through a polar S_N2 mechanism. However, this strategy is limited to molecules containing nucleophilic target atoms. Inspired by a subset of enzymes within the radical SAM superfamily that can methylate inert carbon atoms, we developed an enzymatic strategy to transfer fluoromethyl groups to unactivated carbon atoms. This strategy leverages the ability of halide methyltransferase to generate a transient fluoromethyl-containing SAM analog. Our studies show that <i>S</i>-adenosyl-<i>L</i>-(fluoromethyl)methionine can undergo reductive cleavage to a 5′-deoxyadenosyl 5′-radical, which initiates radical-dependent fluoromethylation through substrate hydrogen-atom abstraction. Adding fluoromethyl groups to unactivated C–H bonds using radical SAM enzymes is a powerful approach that can be used to derivatize molecules of interest where S_N2-based fluoromethylation is precluded.</p>","PeriodicalId":29802,"journal":{"name":"ACS Bio & Med Chem Au","volume":"5 3","pages":"464–474 464–474"},"PeriodicalIF":3.8,"publicationDate":"2025-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://pubs.acs.org/doi/epdf/10.1021/acsbiomedchemau.5c00062","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144305830","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Radical Fluoromethylation Enabled by Cobalamin-Dependent Radical SAM Enzymes.","authors":"Syam Sundar Neti, Bo Wang, Jiayuan Cui, David F Iwig, Nicholas J York, Anthony J Blaszczyk, Matthew R Bauerle, Squire J Booker","doi":"10.1021/acsbiomedchemau.5c00062","DOIUrl":"10.1021/acsbiomedchemau.5c00062","url":null,"abstract":"<p><p>Fluorine is an important atom in many drugs because it can improve the efficacy and metabolic stability of many molecules. Strategies to incorporate monofluoromethyl groups in drugs have been limited and have received less attention than strategies for difluoromethylation or trifluoromethylation. Previously, we and others reported the enzymatic monofluoromethylation of several biologically relevant metabolites based on the transfer of a fluoromethyl group from analogs of <i>S</i>-adenosylmethionine (SAM) to various nucleophiles (carbon, oxygen, nitrogen, sulfur, and carbon) through a polar S_N2 mechanism. However, this strategy is limited to molecules containing nucleophilic target atoms. Inspired by a subset of enzymes within the radical SAM superfamily that can methylate inert carbon atoms, we developed an enzymatic strategy to transfer fluoromethyl groups to unactivated carbon atoms. This strategy leverages the ability of halide methyltransferase to generate a transient fluoromethyl-containing SAM analog. Our studies show that <i>S</i>-adenosyl-<i>L</i>-(fluoromethyl)-methionine can undergo reductive cleavage to a 5'-deoxyadenosyl 5'-radical, which initiates radical-dependent fluoromethylation through substrate hydrogen-atom abstraction. Adding fluoromethyl groups to unactivated C-H bonds using radical SAM enzymes is a powerful approach that can be used to derivatize molecules of interest where S_N2-based fluoromethylation is precluded.</p>","PeriodicalId":29802,"journal":{"name":"ACS Bio & Med Chem Au","volume":"5 3","pages":"464-474"},"PeriodicalIF":3.8,"publicationDate":"2025-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12183590/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144486247","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Mechanism of Forced-Copy-Choice RNA Recombination by Enteroviral RNA-Dependent RNA Polymerases.","authors":"Jamie J Arnold, Alexandre Martinez, Abha Jain, Xinran Liu, Ibrahim M Moustafa, Craig E Cameron","doi":"10.1021/acsbiomedchemau.5c00049","DOIUrl":"10.1021/acsbiomedchemau.5c00049","url":null,"abstract":"<p><p>Forced-copy-choice recombination occurs at the end of a template, differing from copy-choice recombination, which happens at internal positions. This mechanism may produce full-length genomes from fragments created by host antiviral responses. Previous studies from our laboratory demonstrated that poliovirus (PV) RNA-dependent RNA polymerase (RdRp) switches to an \"acceptor\" template <i>in vitro</i> when initiated on a heteropolymeric RNA-primed \"donor\" template. Surprisingly, recombinants showed template switching from the 3'-end of the donor template. We have developed a primed-template system to study PV RdRp-catalyzed forced-copy-choice RNA recombination. PV RdRp adds a single, nontemplated nucleotide to the 3'-end of a blunt-ended, double-stranded RNA product, forming a \"plus-one\" intermediate essential for template switching. Nontemplated addition of CMP was favored over AMP and GMP (80:20:1); UMP addition was negligible. A single basepair between the plus-one intermediate and the 3'-end of the acceptor template was necessary and sufficient for template switching, which could occur without RdRp dissociation. Formation of the plus-one intermediate was rate limiting for template switching. PV RdRp also utilized synthetic, preformed intermediates, including those with UMP 3'-overhangs. Reactions showed up to five consecutive template-switching events, consistent with a repair function for this form of recombination. PV RdRp may exclude UMP during forced-copy-choice RNA recombination to preclude creation of nonsense mutations during RNA fragment assembly. Several other picornaviral RdRps were evaluated, and all were capable of RNA fragment assembly to some extent. Lastly, we propose a structure-based hypothesis for the PV RdRp-plus-one intermediate complex based on an elongating PV RdRp structure.</p>","PeriodicalId":29802,"journal":{"name":"ACS Bio & Med Chem Au","volume":"5 3","pages":"427-446"},"PeriodicalIF":3.8,"publicationDate":"2025-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12183593/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144486244","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Mechanism of Forced-Copy-Choice RNA Recombination by Enteroviral RNA-Dependent RNA Polymerases","authors":"Jamie J. Arnold*,&nbsp;Alexandre Martinez,&nbsp;Abha Jain,&nbsp;Xinran Liu,&nbsp;Ibrahim M. Moustafa and Craig E. Cameron*,&nbsp;","doi":"10.1021/acsbiomedchemau.5c0004910.1021/acsbiomedchemau.5c00049","DOIUrl":"https://doi.org/10.1021/acsbiomedchemau.5c00049https://doi.org/10.1021/acsbiomedchemau.5c00049","url":null,"abstract":"<p >Forced-copy-choice recombination occurs at the end of a template, differing from copy-choice recombination, which happens at internal positions. This mechanism may produce full-length genomes from fragments created by host antiviral responses. Previous studies from our laboratory demonstrated that poliovirus (PV) RNA-dependent RNA polymerase (RdRp) switches to an “acceptor” template <i>in vitro</i> when initiated on a heteropolymeric RNA-primed “donor” template. Surprisingly, recombinants showed template switching from the 3′-end of the donor template. We have developed a primed-template system to study PV RdRp-catalyzed forced-copy-choice RNA recombination. PV RdRp adds a single, nontemplated nucleotide to the 3′-end of a blunt-ended, double-stranded RNA product, forming a “plus-one” intermediate essential for template switching. Nontemplated addition of CMP was favored over AMP and GMP (80:20:1); UMP addition was negligible. A single basepair between the plus-one intermediate and the 3′-end of the acceptor template was necessary and sufficient for template switching, which could occur without RdRp dissociation. Formation of the plus-one intermediate was rate limiting for template switching. PV RdRp also utilized synthetic, preformed intermediates, including those with UMP 3′-overhangs. Reactions showed up to five consecutive template-switching events, consistent with a repair function for this form of recombination. PV RdRp may exclude UMP during forced-copy-choice RNA recombination to preclude creation of nonsense mutations during RNA fragment assembly. Several other picornaviral RdRps were evaluated, and all were capable of RNA fragment assembly to some extent. Lastly, we propose a structure-based hypothesis for the PV RdRp-plus-one intermediate complex based on an elongating PV RdRp structure.</p>","PeriodicalId":29802,"journal":{"name":"ACS Bio & Med Chem Au","volume":"5 3","pages":"427–446 427–446"},"PeriodicalIF":3.8,"publicationDate":"2025-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://pubs.acs.org/doi/epdf/10.1021/acsbiomedchemau.5c00049","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144305829","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"PLAIG: Protein-Ligand Binding Affinity Prediction Using a Novel Interaction-Based Graph Neural Network Framework.","authors":"Madhav V Samudrala, Somanath Dandibhotla, Arjun Kaneriya, Sivanesan Dakshanamurthy","doi":"10.1021/acsbiomedchemau.5c00053","DOIUrl":"10.1021/acsbiomedchemau.5c00053","url":null,"abstract":"<p><p>Rapid prediction of protein-ligand binding affinity is important in the drug discovery process. The advent of machine learning methods has increased the speed of these predictions. Previous machine learning models based on structural, sequence, and interaction-based approaches have shown potential but often tend to memorize training data due to incomplete feature representations that lead to poor generalization on external complexes. To address this challenge, here, we developed PLAIG, a Graph Neural Network (GNN)-based machine learning framework for generalized binding affinity prediction. PLAIG represents binding complexes as graphs, integrating protein-ligand interactions and molecular topology to uniquely capture interaction and structural features. To reduce overfitting, we tested principal component analysis (PCA) and ensemble learning with a stacking regressor. During benchmarking, PLAIG achieved a PCC of 0.78 on 4852 complexes from the PDBbind v.2019 refined set and 0.82 on 285 complexes from the v.2016 core set, outperforming many existing models. External validation on the DUDE-Z data set demonstrated its ability to differentiate active ligands from decoys, achieving an average AUC of 0.69 and a maximum AUC of 0.89. To enrich de novo prediction capabilities for subsequent model versions, PLAIG was hybridized with sequence- and structure-based models. The hybrid models achieved an average PCC of 0.88 on well-known drug-target complexes, with the best reaching a PCC of 0.98. Future work will incorporate an explicit inclusion of a docking methodology into PLAIG's pipeline and assess its performance on de novo ligands. PLAIG is freely available at https://plaig-demo.streamlit.app/.</p>","PeriodicalId":29802,"journal":{"name":"ACS Bio & Med Chem Au","volume":"5 3","pages":"447-463"},"PeriodicalIF":3.8,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12183606/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144486245","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"PLAIG: Protein–Ligand Binding Affinity Prediction Using a Novel Interaction-Based Graph Neural Network Framework","authors":"Madhav V. Samudrala,&nbsp;Somanath Dandibhotla,&nbsp;Arjun Kaneriya and Sivanesan Dakshanamurthy*,&nbsp;","doi":"10.1021/acsbiomedchemau.5c0005310.1021/acsbiomedchemau.5c00053","DOIUrl":"https://doi.org/10.1021/acsbiomedchemau.5c00053https://doi.org/10.1021/acsbiomedchemau.5c00053","url":null,"abstract":"<p >Rapid prediction of protein–ligand binding affinity is important in the drug discovery process. The advent of machine learning methods has increased the speed of these predictions. Previous machine learning models based on structural, sequence, and interaction-based approaches have shown potential but often tend to memorize training data due to incomplete feature representations that lead to poor generalization on external complexes. To address this challenge, here, we developed PLAIG, a Graph Neural Network (GNN)-based machine learning framework for generalized binding affinity prediction. PLAIG represents binding complexes as graphs, integrating protein–ligand interactions and molecular topology to uniquely capture interaction and structural features. To reduce overfitting, we tested principal component analysis (PCA) and ensemble learning with a stacking regressor. During benchmarking, PLAIG achieved a PCC of 0.78 on 4852 complexes from the PDBbind v.2019 refined set and 0.82 on 285 complexes from the v.2016 core set, outperforming many existing models. External validation on the DUDE-Z data set demonstrated its ability to differentiate active ligands from decoys, achieving an average AUC of 0.69 and a maximum AUC of 0.89. To enrich de novo prediction capabilities for subsequent model versions, PLAIG was hybridized with sequence- and structure-based models. The hybrid models achieved an average PCC of 0.88 on well-known drug–target complexes, with the best reaching a PCC of 0.98. Future work will incorporate an explicit inclusion of a docking methodology into PLAIG’s pipeline and assess its performance on de novo ligands. PLAIG is freely available at https://plaig-demo.streamlit.app/.</p>","PeriodicalId":29802,"journal":{"name":"ACS Bio & Med Chem Au","volume":"5 3","pages":"447–463 447–463"},"PeriodicalIF":3.8,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://pubs.acs.org/doi/epdf/10.1021/acsbiomedchemau.5c00053","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144305834","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}