MELLODDY: Cross-pharma Federated Learning at Unprecedented Scale Unlocks Benefits in QSAR without Compromising Proprietary Information.

Federated multipartner machine learning has been touted as an appealing and efficient method to increase the effective training data volume and thereby the predictivity of models, particularly when the generation of training data is resource-intensive. In the landmark MELLODDY project, indeed, each of ten pharmaceutical companies realized aggregated improvements on its own classification or regression models through federated learning. To this end, they leveraged a novel implementation extending multitask learning across partners, on a platform audited for privacy and security. The experiments involved an unprecedented cross-pharma data set of 2.6+ billion confidential experimental activity data points, documenting 21+ million physical small molecules and 40+ thousand assays in on-target and secondary pharmacodynamics and pharmacokinetics. Appropriate complementary metrics were developed to evaluate the predictive performance in the federated setting. In addition to predictive performance increases in labeled space, the results point toward an extended applicability domain in federated learning. Increases in collective training data volume, including by means of auxiliary data resulting from single concentration high-throughput and imaging assays, continued to boost predictive performance, albeit with a saturating return. Markedly higher improvements were observed for the pharmacokinetics and safety panel assay-based task subsets.

Continuous Monitoring of Patient Mobility for 18 Months Using Inertial Sensors following Traumatic Knee Injury: A Case Study.

Continuous patient activity monitoring during rehabilitation, enabled by digital technologies, will allow the objective capture of real-world mobility and aligning treatment to each individual's recovery trajectory in real time. To explore the feasibility and added value of such approaches, we present a case study of a 36-year-old male participant monitored continuously for activity levels and gait parameters using a waist-worn inertial sensor following a tibial plateau fracture on the right side, sustained as a result of a high-energy trauma during a sporting accident. During rehabilitation, data were collected for a period of 553 days, with > 80% daytime compliance, until the participant returned to near full mobility. The participant completed a daily diary with the annotation of major events (falls, near falls, cycling periods, or physiotherapy sessions) and key dates in the patient's recovery, including medical interventions, transitioning off crutches, and returning to work. We demonstrate the feasibility of collecting, storing, and mining of continuous digital mobility data and show that such data can detect changes in mobility and provide insights into long-term rehabilitation. We make both raw data and annotations available as a resource with the aspiration that further methods and insights will be built on this initial exploration of added value and continue to demonstrate that continuous monitoring can be deployed to aid rehabilitation.

Jenkins-CI, an Open-Source Continuous Integration System, as a Scientific Data and Image-Processing Platform.

High-throughput screening generates large volumes of heterogeneous data that require a diverse set of computational tools for management, processing, and analysis. Building integrated, scalable, and robust computational workflows for such applications is challenging but highly valuable. Scientific data integration and pipelining facilitate standardized data processing, collaboration, and reuse of best practices. We describe how Jenkins-CI, an "off-the-shelf," open-source, continuous integration system, is used to build pipelines for processing images and associated data from high-content screening (HCS). Jenkins-CI provides numerous plugins for standard compute tasks, and its design allows the quick integration of external scientific applications. Using Jenkins-CI, we integrated CellProfiler, an open-source image-processing platform, with various HCS utilities and a high-performance Linux cluster. The platform is web-accessible, facilitates access and sharing of high-performance compute resources, and automates previously cumbersome data and image-processing tasks. Imaging pipelines developed using the desktop CellProfiler client can be managed and shared through a centralized Jenkins-CI repository. Pipelines and managed data are annotated to facilitate collaboration and reuse. Limitations with Jenkins-CI (primarily around the user interface) were addressed through the selection of helper plugins from the Jenkins-CI community.

A Western diet ecological module identified from the 'humanized' mouse microbiota predicts diet in adults and formula feeding in children.

The interplay between diet and the microbiota has been implicated in the growing frequency of chronic diseases associated with the Western lifestyle. However, the complexity and variability of microbial ecology in humans and preclinical models has hampered identification of the molecular mechanisms underlying the association of the microbiota in this context. We sought to address two key questions. Can the microbial ecology of preclinical models predict human populations? And can we identify underlying principles that surpass the plasticity of microbial ecology in humans? To do this, we focused our study on diet; perhaps the most influential factor determining the composition of the gut microbiota. Beginning with a study in 'humanized' mice we identified an interactive module of 9 genera allied with Western diet intake. This module was applied to a controlled dietary study in humans. The abundance of the Western ecological module correctly predicted the dietary intake of 19/21 top and 21/21 of the bottom quartile samples inclusive of all 5 Western and 'low-fat' diet subjects, respectively. In 98 volunteers the abundance of the Western module correlated appropriately with dietary intake of saturated fatty acids, fat-soluble vitamins and fiber. Furthermore, it correlated with the geographical location and dietary habits of healthy adults from the Western, developing and third world. The module was also coupled to dietary intake in children (and piglets) correlating with formula (vs breast) feeding and associated with a precipitous development of the ecological module in young children. Our study provides a conceptual platform to translate microbial ecology from preclinical models to humans and identifies an ecological network module underlying the association of the gut microbiota with Western dietary habits.

Host genetics and environmental factors regulate ecological succession of the mouse colon tissue-associated microbiota.

BACKGROUND: The integration of host genetics, environmental triggers and the microbiota is a recognised factor in the pathogenesis of barrier function diseases such as IBD. In order to determine how these factors interact to regulate the host immune response and ecological succession of the colon tissue-associated microbiota, we investigated the temporal interaction between the microbiota and the host following disruption of the colonic epithelial barrier. METHODOLOGY/PRINCIPAL FINDINGS: Oral administration of DSS was applied as a mechanistic model of environmental damage of the colon and the resulting inflammation characterized for various parameters over time in WT and Nod2 KO mice. RESULTS: In WT mice, DSS damage exposed the host to the commensal flora and led to a migration of the tissue-associated bacteria from the epithelium to mucosal and submucosal layers correlating with changes in proinflammatory cytokine profiles and a progressive transition from acute to chronic inflammation of the colon. Tissue-associated bacteria levels peaked at day 21 post-DSS and declined thereafter, correlating with recruitment of innate immune cells and development of the adaptive immune response. Histological parameters, immune cell infiltration and cytokine biomarkers of inflammation were indistinguishable between Nod2 and WT littermates following DSS, however, Nod2 KO mice demonstrated significantly higher tissue-associated bacterial levels in the colon. DSS damage and Nod2 genotype independently regulated the community structure of the colon microbiota. CONCLUSIONS/SIGNIFICANCE: The results of these experiments demonstrate the integration of environmental and genetic factors in the ecological succession of the commensal flora in mammalian tissue. The association of Nod2 genotype (and other host polymorphisms) and environmental factors likely combine to influence the ecological succession of the tissue-associated microflora accounting in part for their association with the pathogenesis of inflammatory bowel diseases.

Cellular cholesterol regulates MT1 MMP dependent activation of MMP 2 via MEK-1 in HT1080 fibrosarcoma cells.

Unstimulated human fibrosarcoma cells (HT1080) constitutively secrete matrix metalloproteinase 2 (MMP 2) as a proenzyme requiring proteolytic cleavage by membrane type-1 MMP (MT1 MMP) for activation. Physiological and pharmacological stimuli induce clustering of MT1 MMP/tissue inhibitor of MMP 2 "receptors", promoting binding and activation of MMP 2. We now report that cholesterol depleted HT1080 cells accumulated MT1 MMP on the cell surface and activated MMP 2. A specific inhibitor of mitogen activated protein kinase kinase 1/2 inhibited both MMP 2 activation and extracellular signal-related kinase phosphorylation induced by cholesterol depletion. Our data indicate that the cholesterol content of unstimulated cells is critical for secretion of MMP 2 as an inactive zymogen and control of pericellular proteolysis.