Pub Date : 2024-09-16DOI: 10.1101/2024.09.12.612623
Marton L. Olbei, Isabelle Hautefort, John P Thomas, Luca L. Csabai, Balazs Bogar, Hajir Ibraheim, Aamir Saifuddin, Dezso Modos, Nick Powell, Tamas Korcsmaros
Ulcerative colitis (UC) is a chronic inflammatory disorder of the gastrointestinal tract characterised by dysregulated cytokine signalling. Despite the advent of advanced therapies targeting cytokine signalling, treatment outcomes for UC patients remain suboptimal. Hence, there is a pressing need to better understand the complexity of cytokine regulation in UC by comprehensively mapping the interconnected cytokine signalling networks that are perturbed in UC patients. To address this, we undertook systems immunology modelling of single-cell transcriptomics data from colonic biopsies of treatment-naive and treatment-exposed UC patients to build complex cytokine signalling networks underpinned by putative cytokine-cytokine interactions. The generated cytokine networks effectively captured known physiologically relevant cytokine-cytokine interactions which we recapitulated in vitro in UC patient-derived colonic epithelial organoids. These networks revealed new aspects of UC pathogenesis, including a cytokine subnetwork that is unique to treatment-naive UC patients, the identification of highly rewired cytokines across UC disease states (IL22, TL1A, IL23A, and OSM), JAK paralogue-specific cytokine-cytokine interactions, and the positioning of TL1A as an important upstream regulator of TNF and IL23A as well as an attractive therapeutic target. Overall, these findings open up several avenues for guiding future cytokine-targeting therapeutic approaches in UC, and the presented methodology can be readily applied to gain similar insights into other immune-mediated inflammatory diseases (IMIDs).
{"title":"Decoding Cytokine Networks in Ulcerative Colitis to Identify Pathogenic Mechanisms and Therapeutic Targets","authors":"Marton L. Olbei, Isabelle Hautefort, John P Thomas, Luca L. Csabai, Balazs Bogar, Hajir Ibraheim, Aamir Saifuddin, Dezso Modos, Nick Powell, Tamas Korcsmaros","doi":"10.1101/2024.09.12.612623","DOIUrl":"https://doi.org/10.1101/2024.09.12.612623","url":null,"abstract":"Ulcerative colitis (UC) is a chronic inflammatory disorder of the gastrointestinal tract characterised by dysregulated cytokine signalling. Despite the advent of advanced therapies targeting cytokine signalling, treatment outcomes for UC patients remain suboptimal. Hence, there is a pressing need to better understand the complexity of cytokine regulation in UC by comprehensively mapping the interconnected cytokine signalling networks that are perturbed in UC patients. To address this, we undertook systems immunology modelling of single-cell transcriptomics data from colonic biopsies of treatment-naive and treatment-exposed UC patients to build complex cytokine signalling networks underpinned by putative cytokine-cytokine interactions. The generated cytokine networks effectively captured known physiologically relevant cytokine-cytokine interactions which we recapitulated in vitro in UC patient-derived colonic epithelial organoids. These networks revealed new aspects of UC pathogenesis, including a cytokine subnetwork that is unique to treatment-naive UC patients, the identification of highly rewired cytokines across UC disease states (IL22, TL1A, IL23A, and OSM), JAK paralogue-specific cytokine-cytokine interactions, and the positioning of TL1A as an important upstream regulator of TNF and IL23A as well as an attractive therapeutic target. Overall, these findings open up several avenues for guiding future cytokine-targeting therapeutic approaches in UC, and the presented methodology can be readily applied to gain similar insights into other immune-mediated inflammatory diseases (IMIDs).","PeriodicalId":501213,"journal":{"name":"bioRxiv - Systems Biology","volume":"202 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142256510","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-16DOI: 10.1101/2024.09.11.612546
Jenna Tomkinson, Cameron Mattson, Michelle Mattson-Hoss, Herb Sarnoff, Stephanie J Bouley, James A Walker, Gregory P Way
Neurofibromatosis type 1 (NF1) is a multi-system, autosomal dominant genetic disorder driven by the systemic loss of the NF1 protein neurofibromin. Loss of neurofibromin in Schwann cells is particularly detrimental, as the acquisition of a second-hit (e.g., complete loss of NF1) can lead to the development of plexiform neurofibroma tumors. Plexiform neurofibromas are painful, disfiguring tumors with an approximately 1 in 5 chance of sarcoma transition. Selumetinib is currently the only medicine approved by the U.S. Food and Drug Administration (FDA) for the treatment of plexiform neurofibromas in a subset of patients. This motivates the need to develop new therapies, either derived to treat NF1 haploinsufficiency or complete loss of NF1 function. To identify new therapies, we need to understand the impact neurofibromin has on Schwann cells. Here, we aimed to characterize differences in high-content microscopy imaging in neurofibromin-deficient Schwann cells. We applied a fluorescence microscopy assay (called Cell Painting) to two isogenic Schwann cell lines, one of wildtype genotype (NF1+/+) and one of NF1 null genotype (NF1-/-). We modified the canonical Cell Painting assay to mark four organelles/subcellular compartments: nuclei, endoplasmic reticulum, mitochondria, and F-actin. We utilized CellProfiler pipelines to perform quality control, illumination correction, segmentation, and cell morphology feature extraction. We segmented 22,585 NF1 wildtype and null cells, utilized 907 significant cell morphology features representing various organelle shapes and intensity patterns, and trained a logistic regression machine learning model to predict the NF1 genotype of single Schwann cells. The machine learning model had high performance, with training and testing data yielding a balanced accuracy of 0.85 and 0.80, respectively. All of our data processing and analyses are freely available on GitHub. We look to improve upon this preliminary model in the future by applying it to large-scale drug screens of NF1 deficient cells to identify candidate drugs that return NF1 patient Schwann cells to phenocopy NF1 wildtype and healthier phenotype.
{"title":"High-content microscopy and machine learning characterize a cell morphology signature of NF1 genotype in Schwann cells","authors":"Jenna Tomkinson, Cameron Mattson, Michelle Mattson-Hoss, Herb Sarnoff, Stephanie J Bouley, James A Walker, Gregory P Way","doi":"10.1101/2024.09.11.612546","DOIUrl":"https://doi.org/10.1101/2024.09.11.612546","url":null,"abstract":"Neurofibromatosis type 1 (NF1) is a multi-system, autosomal dominant genetic disorder driven by the systemic loss of the NF1 protein neurofibromin. Loss of neurofibromin in Schwann cells is particularly detrimental, as the acquisition of a second-hit (e.g., complete loss of NF1) can lead to the development of plexiform neurofibroma tumors. Plexiform neurofibromas are painful, disfiguring tumors with an approximately 1 in 5 chance of sarcoma transition. Selumetinib is currently the only medicine approved by the U.S. Food and Drug Administration (FDA) for the treatment of plexiform neurofibromas in a subset of patients. This motivates the need to develop new therapies, either derived to treat NF1 haploinsufficiency or complete loss of NF1 function. To identify new therapies, we need to understand the impact neurofibromin has on Schwann cells. Here, we aimed to characterize differences in high-content microscopy imaging in neurofibromin-deficient Schwann cells. We applied a fluorescence microscopy assay (called Cell Painting) to two isogenic Schwann cell lines, one of wildtype genotype (NF1+/+) and one of NF1 null genotype (NF1-/-). We modified the canonical Cell Painting assay to mark four organelles/subcellular compartments: nuclei, endoplasmic reticulum, mitochondria, and F-actin. We utilized CellProfiler pipelines to perform quality control, illumination correction, segmentation, and cell morphology feature extraction. We segmented 22,585 NF1 wildtype and null cells, utilized 907 significant cell morphology features representing various organelle shapes and intensity patterns, and trained a logistic regression machine learning model to predict the NF1 genotype of single Schwann cells. The machine learning model had high performance, with training and testing data yielding a balanced accuracy of 0.85 and 0.80, respectively. All of our data processing and analyses are freely available on GitHub. We look to improve upon this preliminary model in the future by applying it to large-scale drug screens of NF1 deficient cells to identify candidate drugs that return NF1 patient Schwann cells to phenocopy NF1 wildtype and healthier phenotype.","PeriodicalId":501213,"journal":{"name":"bioRxiv - Systems Biology","volume":"3 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142256511","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-16DOI: 10.1101/2024.09.11.612537
Rene Coig, Benjamin R Harrison, Richard S Johnson, Michael J MacCoss, Daniel EL Promislow
Sex has a major effect on the metabolome. However, we do not yet understand the degree to which these quantitative sex differences in metabolism are associated with anatomical dimorphism and modulated by sex-specific tissues. In the fruit fly, Drosophila melanogaster, knocking out the doublesex (dsx) gene gives rise to adults with intermediate sex characteristics. Here we sought to determine the degree to which this key node in sexual development leads to sex differences in the fly metabolome. We measured 91 metabolites across head, thorax and abdomen in Drosophila, comparing the differences between distinctly sex-dimorphic flies with those of reduced sexual dimorphism: dsx null flies. Notably, in the reduced dimorphism flies, we observed a sex difference in only 1 of 91 metabolites, kynurenate, whereas 51% of metabolites (46/91) were significantly different between wildtype XX and XY flies in at least one tissue, suggesting that dsx plays a major role in sex differences in fly metabolism. Kynurenate was consistently higher in XX flies in both the presence and absence of functioning dsx. We observed tissue-specific consequences of knocking out dsx. Metabolites affected by sex were significantly enriched in branched chain amino acid metabolism and the mTOR pathway. This highlights the importance of considering variation in genes that cause anatomical sexual dimorphism when analyzing sex differences in metabolic profiles and interpreting their biological significance.
{"title":"Tissue-specific metabolomic signatures for a doublesex model of reduced sexual dimorphism","authors":"Rene Coig, Benjamin R Harrison, Richard S Johnson, Michael J MacCoss, Daniel EL Promislow","doi":"10.1101/2024.09.11.612537","DOIUrl":"https://doi.org/10.1101/2024.09.11.612537","url":null,"abstract":"Sex has a major effect on the metabolome. However, we do not yet understand the degree to which these quantitative sex differences in metabolism are associated with anatomical dimorphism and modulated by sex-specific tissues. In the fruit fly, Drosophila melanogaster, knocking out the doublesex (dsx) gene gives rise to adults with intermediate sex characteristics. Here we sought to determine the degree to which this key node in sexual development leads to sex differences in the fly metabolome. We measured 91 metabolites across head, thorax and abdomen in Drosophila, comparing the differences between distinctly sex-dimorphic flies with those of reduced sexual dimorphism: dsx null flies. Notably, in the reduced dimorphism flies, we observed a sex difference in only 1 of 91 metabolites, kynurenate, whereas 51% of metabolites (46/91) were significantly different between wildtype XX and XY flies in at least one tissue, suggesting that dsx plays a major role in sex differences in fly metabolism. Kynurenate was consistently higher in XX flies in both the presence and absence of functioning dsx. We observed tissue-specific consequences of knocking out dsx. Metabolites affected by sex were significantly enriched in branched chain amino acid metabolism and the mTOR pathway. This highlights the importance of considering variation in genes that cause anatomical sexual dimorphism when analyzing sex differences in metabolic profiles and interpreting their biological significance.","PeriodicalId":501213,"journal":{"name":"bioRxiv - Systems Biology","volume":"31 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142256513","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-14DOI: 10.1101/2024.09.09.612037
Shubham Gajrani, Xiaozhou Ye, Christoph Ratzke
Microbes usually live in complex communities interacting with many other microbial species. These interactions determine who can persist in a community and how the overall community forms and functions. Microbes often exert interactions by chemically changing the environment, like taking up nutrients or producing toxins. These environmental changes can persist over time. We show here that such lasting environmental changes can cause a memory effect where current growth conditions alter interaction outcomes in the future. Importantly, this memory is only stored in the environment and not inside the bacterial cells. Only the collective effort of many bacteria can build up this memory, making it an emergent property of bacterial populations. This population memory can also impact the assembly of more complex communities and lead to different final communities depending on the system's past. Overall, we show that to understand interaction outcomes fully, we not only have to consider the interacting species and abiotic conditions but also the system's history.
{"title":"Environment-mediated interactions cause an externalized and collective memory in microbes","authors":"Shubham Gajrani, Xiaozhou Ye, Christoph Ratzke","doi":"10.1101/2024.09.09.612037","DOIUrl":"https://doi.org/10.1101/2024.09.09.612037","url":null,"abstract":"Microbes usually live in complex communities interacting with many other microbial species. These interactions determine who can persist in a community and how the overall community forms and functions. Microbes often exert interactions by chemically changing the environment, like taking up nutrients or producing toxins. These environmental changes can persist over time. We show here that such lasting environmental changes can cause a memory effect where current growth conditions alter interaction outcomes in the future. Importantly, this memory is only stored in the environment and not inside the bacterial cells. Only the collective effort of many bacteria can build up this memory, making it an emergent property of bacterial populations. This population memory can also impact the assembly of more complex communities and lead to different final communities depending on the system's past. Overall, we show that to understand interaction outcomes fully, we not only have to consider the interacting species and abiotic conditions but also the system's history.","PeriodicalId":501213,"journal":{"name":"bioRxiv - Systems Biology","volume":"13 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142256515","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-14DOI: 10.1101/2024.09.13.610890
EMMANUEL ORTEGA ATEHORTUA, Juan Camilo Arboleda, Nicole Rivera, Gloria Machado, Boris Anghelo Rodriguez
Giant viruses have been in the scope of virologists since 2003 when they were isolated from Acanthamoeba spp. Giant viruses, in turn, get infected by another virus named virophage and a third biological entity that corresponds to a transpoviron which can be found in the capsids of giant and virophage viruses. So far, transpovirons seem to behave as commensal entities while some virophages exhibit commensal behavior under laboratory conditions. To study the system's behavior, we used a theoretical approximation and developed an ordinary differential equation model. The dynamical analysis showed that the system exhibits an oscillatory robust behavior leading to a hyperparasitic Lotka-Volterra dynamic. But the biological mechanism that underlines the transpoviron persistence over time remains unclear and its status as a commensal entity needs further assessment. Also, the ecological interaction that leads to the overall coexistence of the three viral entities needs to be further studied.
{"title":"Parasitic and Commensal interactions among Mimiviruses, Sputnik-like virophages, and Transpovirons: A theoretical and dynamical systems approach.","authors":"EMMANUEL ORTEGA ATEHORTUA, Juan Camilo Arboleda, Nicole Rivera, Gloria Machado, Boris Anghelo Rodriguez","doi":"10.1101/2024.09.13.610890","DOIUrl":"https://doi.org/10.1101/2024.09.13.610890","url":null,"abstract":"Giant viruses have been in the scope of virologists since 2003 when they were isolated from <em>Acanthamoeba</em> spp. Giant viruses, in turn, get infected by another virus named virophage and a third biological entity that corresponds to a transpoviron which can be found in the capsids of giant and virophage viruses. So far, transpovirons seem to behave as commensal entities while some virophages exhibit commensal behavior under laboratory conditions. To study the system's behavior, we used a theoretical approximation and developed an ordinary differential equation model. The dynamical analysis showed that the system exhibits an oscillatory robust behavior leading to a hyperparasitic Lotka-Volterra dynamic. But the biological mechanism that underlines the transpoviron persistence over time remains unclear and its status as a commensal entity needs further assessment. Also, the ecological interaction that leads to the overall coexistence of the three viral entities needs to be further studied.","PeriodicalId":501213,"journal":{"name":"bioRxiv - Systems Biology","volume":"72 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142256517","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-14DOI: 10.1101/2024.09.12.612709
Joshua Cook, Eric Ron, Dmitri Svetlov, Luis Aguiulera, Brian Munsky
Control of gene regulation requires quantitatively accurate predictions of heterogeneous cellular responses. When inferred from single-cell experiments, discrete stochastic models can enable such predictions, but such experiments are highly adjustable, allowing for almost infinitely many potential designs (e.g., at different induction levels, for different measurement times, or considering different observed biological species). Not all experiments are equally informative, experiments are time-consuming or expensive to perform, and research begins with limited prior information with which to construct models. To address these concerns, we developed a sequential experiment design strategy that starts with simple preliminary experiments and then integrates chemical master equations to compute the likelihood of single-cell data, a Bayesian inference procedure to sample posterior parameter distributions, and a finite state projection based Fisher information matrix to estimate the expected information for different designs for subsequent experiments. Using simulated then real single-cell data, we determined practical working principles to reduce the overall number of experiments needed to achieve predictive, quantitative understanding of single-cell responses.
{"title":"Sequential design of single-cell experiments to identify discrete stochastic models for gene expression.","authors":"Joshua Cook, Eric Ron, Dmitri Svetlov, Luis Aguiulera, Brian Munsky","doi":"10.1101/2024.09.12.612709","DOIUrl":"https://doi.org/10.1101/2024.09.12.612709","url":null,"abstract":"Control of gene regulation requires quantitatively accurate predictions of heterogeneous cellular responses. When inferred from single-cell experiments, discrete stochastic models can enable such predictions, but such experiments are highly adjustable, allowing for almost infinitely many potential designs (e.g., at different induction levels, for different measurement times, or considering different observed biological species). Not all experiments are equally informative, experiments are time-consuming or expensive to perform, and research begins with limited prior information with which to construct models. To address these concerns, we developed a sequential experiment design strategy that starts with simple preliminary experiments and then integrates chemical master equations to compute the likelihood of single-cell data, a Bayesian inference procedure to sample posterior parameter distributions, and a finite state projection based Fisher information matrix to estimate the expected information for different designs for subsequent experiments. Using simulated then real single-cell data, we determined practical working principles to reduce the overall number of experiments needed to achieve predictive, quantitative understanding of single-cell responses.","PeriodicalId":501213,"journal":{"name":"bioRxiv - Systems Biology","volume":"16 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142256512","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-13DOI: 10.1101/2024.09.09.612059
Sam Ganzfried
Recent work by Kleshnina et al. has presented a Stackelberg evolutionary game model in which the Stackelberg equilibrium strategy for the leading player corresponds to the optimal cancer treatment. We present an approach that is able to quickly and accurately solve the model presented in that work.
{"title":"Computing Stackelberg Equilibrium for Cancer Treatment","authors":"Sam Ganzfried","doi":"10.1101/2024.09.09.612059","DOIUrl":"https://doi.org/10.1101/2024.09.09.612059","url":null,"abstract":"Recent work by Kleshnina et al. has presented a Stackelberg evolutionary game model in which the Stackelberg equilibrium strategy for the leading player corresponds to the optimal cancer treatment. We present an approach that is able to quickly and accurately solve the model presented in that work.","PeriodicalId":501213,"journal":{"name":"bioRxiv - Systems Biology","volume":"53 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142256514","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Despite the growing interest in anti-aging drug development, high cost and low success rate pose a significant challenge. We present ElixirSeeker, a new machine-learning framework designed to help speed up the discovery of potential anti-aging compounds by utilizing the attention-driven fusion of molecular fingerprints. Our approach integrates molecular fingerprints generated by different algorithms and utilizes XGBoost to select optimal fingerprint lengths. Subsequently, we assign weights to the molecular fingerprints and employ Kernel Principal Component Analysis (KPCA) to reduce dimensionality, integrating different attention-driven methods. We trained the algorithm using DrugAge database. Our comprehensive analyses demonstrate that 64-bit Attention-ElixirFP maintains high predictive accuracy and F1 score while minimizing computational cost. Using ElixirSeeker to screen external compound databases, we identified a number of promising candidate anti-aging drugs. We tested top 6 hits and found that 4 of these compounds extend the lifespan of Caenorhabditis elegans, including Polyphyllin Ⅵ, Medrysone, Thymoquinone and Medrysone. This study illustrates that attention-driven fusion of fingerprints maximizes the learning of molecular activity features, providing a novel approach for high-throughput machine learning discovery of anti-aging molecules.
{"title":"ElixirSeeker: A Machine Learning Framework Utilizing Attention-Driven Fusion of Molecular Fingerprints for the Discovery of Anti-Aging Compounds","authors":"Yan Pan, Hongxia Cai, Fang Ye, Wentao Xu, Zhihang Huang, Jingyuan Zhu, Yiwen Gong, Yutong Li, Anastasia Ngozi Ezemaduka, Shan Gao, Shunqi Liu, Guojun Li, Hao Li, Jing Yang, Junyu Ning, Bo Xian","doi":"10.1101/2024.09.08.611839","DOIUrl":"https://doi.org/10.1101/2024.09.08.611839","url":null,"abstract":"Despite the growing interest in anti-aging drug development, high cost and low success rate pose a significant challenge. We present ElixirSeeker, a new machine-learning framework designed to help speed up the discovery of potential anti-aging compounds by utilizing the attention-driven fusion of molecular fingerprints. Our approach integrates molecular fingerprints generated by different algorithms and utilizes XGBoost to select optimal fingerprint lengths. Subsequently, we assign weights to the molecular fingerprints and employ Kernel Principal Component Analysis (KPCA) to reduce dimensionality, integrating different attention-driven methods. We trained the algorithm using DrugAge database. Our comprehensive analyses demonstrate that 64-bit Attention-ElixirFP maintains high predictive accuracy and F1 score while minimizing computational cost. Using ElixirSeeker to screen external compound databases, we identified a number of promising candidate anti-aging drugs. We tested top 6 hits and found that 4 of these compounds extend the lifespan of Caenorhabditis elegans, including Polyphyllin Ⅵ, Medrysone, Thymoquinone and Medrysone. This study illustrates that attention-driven fusion of fingerprints maximizes the learning of molecular activity features, providing a novel approach for high-throughput machine learning discovery of anti-aging molecules.","PeriodicalId":501213,"journal":{"name":"bioRxiv - Systems Biology","volume":"8 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142256642","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-13DOI: 10.1101/2024.09.09.609906
Luke S Ferro, Alan Y. L. Wong, Jack Howland, Ana S. H. Costa, Jefferson G. Pruyne, Devesh Shah, Joshua D. Lauterbach, Steven B. Hooper, Mimoun Cadosch Delmar, Jack Geremia, Timothy Kassis, Naama Kanarek, Jennifer M. Campbell
Mass spectrometry-based metabolomics allows for the quantitation of metabolite levels in diverse biological samples. The traditional method of converting peak areas to absolute concentrations involves the use of matched heavy isotopologues. However, this approach is laborious and limited to a small number of metabolites. We addressed these limitations by developing PyxisTM, a machine learning-based technology which converts raw mass spectrometry data to absolute concentration measurements without the need for per-analyte standards. Here, we demonstrate Pyxis performance by quantifying metabolome concentration dynamics in murine blood plasma. Pyxis performed equivalently to traditional quantitation workflows used by research institutions, with a fraction of the time needed for analysis. We show that absolute quantitation by Pyxis can be expanded to include concentrations for additional metabolites, without the need to acquire new data. Furthermore, Pyxis allows for absolute quantitation as part of an untargeted metabolomics workflow. By removing the bottleneck of per-analyte standards, Pyxis allows for absolute quantitation in metabolomics that is scalable to large numbers of metabolites. The ability of Pyxis to make concentration-based measurements across the metabolome has the potential to deepen our understanding of diverse metabolic perturbations.
{"title":"A scalable approach to absolute quantitation in metabolomics","authors":"Luke S Ferro, Alan Y. L. Wong, Jack Howland, Ana S. H. Costa, Jefferson G. Pruyne, Devesh Shah, Joshua D. Lauterbach, Steven B. Hooper, Mimoun Cadosch Delmar, Jack Geremia, Timothy Kassis, Naama Kanarek, Jennifer M. Campbell","doi":"10.1101/2024.09.09.609906","DOIUrl":"https://doi.org/10.1101/2024.09.09.609906","url":null,"abstract":"Mass spectrometry-based metabolomics allows for the quantitation of metabolite levels in diverse biological samples. The traditional method of converting peak areas to absolute concentrations involves the use of matched heavy isotopologues. However, this approach is laborious and limited to a small number of metabolites. We addressed these limitations by developing PyxisTM, a machine learning-based technology which converts raw mass spectrometry data to absolute concentration measurements without the need for per-analyte standards. Here, we demonstrate Pyxis performance by quantifying metabolome concentration dynamics in murine blood plasma. Pyxis performed equivalently to traditional quantitation workflows used by research institutions, with a fraction of the time needed for analysis. We show that absolute quantitation by Pyxis can be expanded to include concentrations for additional metabolites, without the need to acquire new data. Furthermore, Pyxis allows for absolute quantitation as part of an untargeted metabolomics workflow. By removing the bottleneck of per-analyte standards, Pyxis allows for absolute quantitation in metabolomics that is scalable to large numbers of metabolites. The ability of Pyxis to make concentration-based measurements across the metabolome has the potential to deepen our understanding of diverse metabolic perturbations.","PeriodicalId":501213,"journal":{"name":"bioRxiv - Systems Biology","volume":"43 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142256641","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-13DOI: 10.1101/2024.09.09.612023
Xiongwen Li, Zhu Liang, Zhetao Guo, Ziyi Liu, Ke Wu, Jiahao Luo, Yuesheng Zhang, Lizheng Liu, Manda Sun, Yuanyuan Huang, Hongting Tang, Yu Chen, Tao Yu, Jens Nielsen, Feiran Li
Establishing efficient cell factories involves a continuous process of trial and error due to the intricate nature of metabolism. This complexity makes predicting effective engineering targets a challenging task. Therefore, it is vital to learn from the accumulated successes of previous designs for advancing future cell factory development. In this study, we developed a method based on large language models (LLMs) to extract metabolic engineering strategies from research articles on a large scale. We created a database containing over 29006 metabolic engineering entries, 1210 products and 751 organisms. Using this extracted data, we trained a hybrid model combining deep learning and mechanistic approaches to predict engineering targets. Our model outperformed traditional metabolic engineering target prediction algorithms, excelled in predicting the effects of gene modifications, and generalized well to out-of-distribution products and multiple gene combinations. Our study provides a valuable dataset, a chatbot, and an engineering target prediction model for the metabolic engineering field and exemplifies an efficient method for leveraging existing knowledge for future predictions.
{"title":"Leveraging large language models for metabolic engineering design","authors":"Xiongwen Li, Zhu Liang, Zhetao Guo, Ziyi Liu, Ke Wu, Jiahao Luo, Yuesheng Zhang, Lizheng Liu, Manda Sun, Yuanyuan Huang, Hongting Tang, Yu Chen, Tao Yu, Jens Nielsen, Feiran Li","doi":"10.1101/2024.09.09.612023","DOIUrl":"https://doi.org/10.1101/2024.09.09.612023","url":null,"abstract":"Establishing efficient cell factories involves a continuous process of trial and error due to the intricate nature of metabolism. This complexity makes predicting effective engineering targets a challenging task. Therefore, it is vital to learn from the accumulated successes of previous designs for advancing future cell factory development. In this study, we developed a method based on large language models (LLMs) to extract metabolic engineering strategies from research articles on a large scale. We created a database containing over 29006 metabolic engineering entries, 1210 products and 751 organisms. Using this extracted data, we trained a hybrid model combining deep learning and mechanistic approaches to predict engineering targets. Our model outperformed traditional metabolic engineering target prediction algorithms, excelled in predicting the effects of gene modifications, and generalized well to out-of-distribution products and multiple gene combinations. Our study provides a valuable dataset, a chatbot, and an engineering target prediction model for the metabolic engineering field and exemplifies an efficient method for leveraging existing knowledge for future predictions.","PeriodicalId":501213,"journal":{"name":"bioRxiv - Systems Biology","volume":"17 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142256516","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：