Periodic Reporting for period 5 - COMBINE (Collaboration for Prevention and Treatmentof MDR Bacterial Infections)
Berichtszeitraum: 2023-11-01 bis 2024-10-31
Antimicrobial resistance (AMR) is the ability of a microorganism (such as bacteria, viruses, and some parasites) to stop an antimicrobial (such as antibiotics, antivirals and antimalarials) from working against it. New resistance mechanisms are emerging and spreading globally, threatening our ability to treat infectious diseases, which results in prolonged illness, disability, and death. The AMR Accelerator was created to respond to the challenge of developing new treatment and prevention approaches to AMR infections. More than 90 partners are involved in the AMR Accelerator in nine different projects. The COMBINE project supports the coordination of the AMR Accelerator projects. Additionally, COMBINE will propose innovative methods to optimize preclinical and clinical research for the development of products against AMR infections.
Coordinating the AMR Accelerator: In Period 5 of the project, COMBINE has continued to coordinate support to the AMR Accelerator projects via cross-project interactions including the Coordination Committee, the Data Management Group, the Ethics Advisory Board, the Communication Advisory Board (CAB), and the Communication Champions Team. COMBINE also organised specific cross-project events to bring the different projects and partners together to promote scientific engagement, to ensure learnings and best practices are shared, and to identify synergies which may drive faster project progress. These events include in person cross-project meeting in March 2024 (with project updates, networking, Scientific Interest Groups (SIG) and SWOT feedback sessions), several virtual meetings focused on individual SIG´s and a COMBINE F2F meeting throughout the year.
The SIGs have been setup to get the most of ongoing research within the AMR Accelerator and to facilitate cross-project panel discussion on topics of interest to all. In Period 5, COMBINE has hosted and moderated 13 SIG meetings. These include the following SIG’s: a) Machine Learning SIG (5); b) Animal Models SIG (2);c) Science Communication SIG (4) and d) two public webinars. These SIG’s create communities of experts and practice from the whole AMR Accelerator and will results in tangible outcomes, including publications and White Papers.
Achieving effective management of scientific data: COMBINE has developed a robust data management process and will continue to improve on this process. The software developed within COMBINE will support a wide range of data formats, from simple office documents and chemical structures to preclinical and clinical data sets, while retaining compatibility with all common operating systems. To facilitate data management, a cloud based electronic lab notebook system has been introduced by COMBINE for collection, analysis, storage and management of results at AMR Accelerator partner sites. This is complemented by a central data repository for sharing, aggregating, integrating and analysing data. Documentation covering data governance, quality, access, and analysis procedures has been developed and shared with the projects within the Accelerator.
Alongside the data management support for AMR Accelerator projects and expertise knowledge graphs, we have developed the AntiMicrobial Knowledge Graph (https://antimicrobial-kg.serve.scilifelab.se/) that servers as a one-stop-shop for small molecule antimicrobial assays. To generate this, we aggregate data from three main bioactivity data resources namely ChEMBL, CO-ADD, and SPARK and systematically harmonize the compounds and activity endpoints across the data sets. This resource is one of the largest antimicrobial dataset present currently with more than 81,000 compounds tested across 1,373 bacterial strain. The group used these data to develop a model to predict the antimicrobial activity of small molecules, aiming to classify compounds as either active or inactive. For active compounds, the model further provides specificity regarding the pathogen class, including Gram-positive, Gram-negative, Acid-fast bacteria, and fungi. Validation of the model was performed using experimental results from two compound libraries: the EU-OPENSCREEN library and the Enamine Antibacterial Library. The models, underlying datasets, and scripts used for training have been deposited on GitHub and Zenodo, with an associated publication currently in preparation (https://github.com/IMI-COMBINE/broad_spectrum_prediction .
Managing the communication: COMBINE works to strengthen the interaction between AMR Accelerator participants and AMR stakeholders across the EU and globally. In Period 5 of the project, COMBINE developed a web resource to showcase the tools and research infrastructures developed by the project. The project has promoted the AMR Accelerator to the external scientific community and the general public through an e-mail newsletter, public webinars, press-releases, social media (X, LinkedIn and YouTube), the AMR Accelerator website. Based on a SWOT analysis from the cross-project meeting in March 2023 and input from all 9 projects, our joint AMR Accelerator paper (on the benefits and challenges faced within the AMR Accelerator) was accepted by Nature Drug Discovery Review Journal in time to have impact at the higher AMR discussion that occurred in 2024. COMBINE developed and coordinated a global publicity campaign for the article, with global AAAS EurekAlert! and AlphaGalileo press releases from Uppsala University, that was mirrored by partners with project-specific messages, resulting in an Altmetric Score of 113.
Improving clinical trials and standardising an animal infection model: In terms of the scientific goals of the project, COMBINE has progressed towards the capability building objectives of the project. With the aim to identify promising strategies to improve translation from preclinical models to successful clinical trials for products against AMR infections, common problems in the preclinical (including animal infection models) and clinical development of vaccines and antibiotics have been collected. In Period 4, COMBINE has successfully acquired individual participant level data from two development programs and initiated the re-analysis of such data. Moreover, we are using information from additional sources (published literature, EMA Scientific Advice letters) to integrate and expand our knowledge about possible pitfalls and mitigation strategies in translation and clinical trials. The developed infection model protocol has been successfully implemented in three different labs using sharable bacterial strains, deposited in a public biorepository. The collaboration with some of the main actors in the field of animal infection models was further established.
The SIGs have been setup to get the most of ongoing research within the AMR Accelerator and to facilitate cross-project panel discussion on topics of interest to all. In Period 5, COMBINE has hosted and moderated 13 SIG meetings. These include the following SIG’s: a) Machine Learning SIG (5); b) Animal Models SIG (2);c) Science Communication SIG (4) and d) two public webinars. These SIG’s create communities of experts and practice from the whole AMR Accelerator and will results in tangible outcomes, including publications and White Papers.
Achieving effective management of scientific data: COMBINE has developed a robust data management process and will continue to improve on this process. The software developed within COMBINE will support a wide range of data formats, from simple office documents and chemical structures to preclinical and clinical data sets, while retaining compatibility with all common operating systems. To facilitate data management, a cloud based electronic lab notebook system has been introduced by COMBINE for collection, analysis, storage and management of results at AMR Accelerator partner sites. This is complemented by a central data repository for sharing, aggregating, integrating and analysing data. Documentation covering data governance, quality, access, and analysis procedures has been developed and shared with the projects within the Accelerator.
Alongside the data management support for AMR Accelerator projects and expertise knowledge graphs, we have developed the AntiMicrobial Knowledge Graph (https://antimicrobial-kg.serve.scilifelab.se/) that servers as a one-stop-shop for small molecule antimicrobial assays. To generate this, we aggregate data from three main bioactivity data resources namely ChEMBL, CO-ADD, and SPARK and systematically harmonize the compounds and activity endpoints across the data sets. This resource is one of the largest antimicrobial dataset present currently with more than 81,000 compounds tested across 1,373 bacterial strain. The group used these data to develop a model to predict the antimicrobial activity of small molecules, aiming to classify compounds as either active or inactive. For active compounds, the model further provides specificity regarding the pathogen class, including Gram-positive, Gram-negative, Acid-fast bacteria, and fungi. Validation of the model was performed using experimental results from two compound libraries: the EU-OPENSCREEN library and the Enamine Antibacterial Library. The models, underlying datasets, and scripts used for training have been deposited on GitHub and Zenodo, with an associated publication currently in preparation (https://github.com/IMI-COMBINE/broad_spectrum_prediction .
Managing the communication: COMBINE works to strengthen the interaction between AMR Accelerator participants and AMR stakeholders across the EU and globally. In Period 5 of the project, COMBINE developed a web resource to showcase the tools and research infrastructures developed by the project. The project has promoted the AMR Accelerator to the external scientific community and the general public through an e-mail newsletter, public webinars, press-releases, social media (X, LinkedIn and YouTube), the AMR Accelerator website. Based on a SWOT analysis from the cross-project meeting in March 2023 and input from all 9 projects, our joint AMR Accelerator paper (on the benefits and challenges faced within the AMR Accelerator) was accepted by Nature Drug Discovery Review Journal in time to have impact at the higher AMR discussion that occurred in 2024. COMBINE developed and coordinated a global publicity campaign for the article, with global AAAS EurekAlert! and AlphaGalileo press releases from Uppsala University, that was mirrored by partners with project-specific messages, resulting in an Altmetric Score of 113.
Improving clinical trials and standardising an animal infection model: In terms of the scientific goals of the project, COMBINE has progressed towards the capability building objectives of the project. With the aim to identify promising strategies to improve translation from preclinical models to successful clinical trials for products against AMR infections, common problems in the preclinical (including animal infection models) and clinical development of vaccines and antibiotics have been collected. In Period 4, COMBINE has successfully acquired individual participant level data from two development programs and initiated the re-analysis of such data. Moreover, we are using information from additional sources (published literature, EMA Scientific Advice letters) to integrate and expand our knowledge about possible pitfalls and mitigation strategies in translation and clinical trials. The developed infection model protocol has been successfully implemented in three different labs using sharable bacterial strains, deposited in a public biorepository. The collaboration with some of the main actors in the field of animal infection models was further established.
Expected results until the end of the project and potential impacts:
Network: A knowledge base and interpersonal network will develop between AMR Accelerator participants and AMR stakeholders across the EU and globally, which will be important for future contacts and potential collaborations in the AMR community. COMBINE will be instrumental in ensuring that these ties will remain.
Improving clinical trials and standardising animal models: COMBINE will contribute to improve design of clinical trials and develop more predictive and reliable infection models for preclinical studies, to accelerate antibacterial drug and vaccine development.
Network: A knowledge base and interpersonal network will develop between AMR Accelerator participants and AMR stakeholders across the EU and globally, which will be important for future contacts and potential collaborations in the AMR community. COMBINE will be instrumental in ensuring that these ties will remain.
Improving clinical trials and standardising animal models: COMBINE will contribute to improve design of clinical trials and develop more predictive and reliable infection models for preclinical studies, to accelerate antibacterial drug and vaccine development.