ABSTRACT
Fast and reliable detection of patients with severe and heterogeneous illnesses is a major goal of precision medicine1,2. Patients with leukaemia can be identified using machine learning on the basis of their blood transcriptomes3. However, there is an increasing divide between what is technically possible and what is allowed, because of privacy legislation4,5. Here, to facilitate the integration of any medical data from any data owner worldwide without violating privacy laws, we introduce Swarm Learning-a decentralized machine-learning approach that unites edge computing, blockchain-based peer-to-peer networking and coordination while maintaining confidentiality without the need for a central coordinator, thereby going beyond federated learning. To illustrate the feasibility of using Swarm Learning to develop disease classifiers using distributed data, we chose four use cases of heterogeneous diseases (COVID-19, tuberculosis, leukaemia and lung pathologies). With more than 16,400 blood transcriptomes derived from 127 clinical studies with non-uniform distributions of cases and controls and substantial study biases, as well as more than 95,000 chest X-ray images, we show that Swarm Learning classifiers outperform those developed at individual sites. In addition, Swarm Learning completely fulfils local confidentiality regulations by design. We believe that this approach will notably accelerate the introduction of precision medicine.
Subject(s)
Blockchain , Clinical Decision-Making/methods , Confidentiality , Datasets as Topic , Machine Learning , Precision Medicine/methods , COVID-19/diagnosis , COVID-19/epidemiology , Disease Outbreaks , Female , Humans , Leukemia/diagnosis , Leukemia/pathology , Leukocytes/pathology , Lung Diseases/diagnosis , Machine Learning/trends , Male , Software , Tuberculosis/diagnosisABSTRACT
Differentiation of inflammatory macrophages from monocytes is characterized by an orderly integration of epigenetic and transcriptional regulatory mechanisms guided by lineage-determining transcription factors such as PU.1. Further activation of macrophages leads to a stimulus- or microenvironment-specific signal integration with subsequent transcriptional control established by the action of tissue- or signal-associated transcription factors. Here, we assess four histone modifications during human macrophage activation and integrate this information with the gene expression data from 28 different macrophage activation conditions in combination with GM-CSF. Bioinformatically, for inflammatory macrophages we define a unique network of transcriptional and epigenetic regulators (TRs), which was characterized by accessible promoters independent of the activation signal. In contrast to the general accessibility of promoters of TRs, mRNA expression of central TRs belonging to the TR network displayed stimulus-specific expression patterns, indicating a second level of transcriptional regulation beyond epigenetic chromatin changes. In contrast, stringent integration of epigenetic and transcriptional regulation was observed in networks of TRs established from somatic tissues and tissue macrophages. In these networks, clusters of TRs with permissive histone marks were associated with high gene expression whereas clusters with repressive chromatin marks were associated with absent gene expression. Collectively, these results support that macrophage activation during inflammation in contrast to lineage determination is mainly regulated transcriptionally by a pre-defined TR network.