13 January 2021 in Genome Medicine
Background: The SARS-CoV-2 pandemic is currently leading to increasing numbers of COVID-19 patients all over the world. Clinical presentations range from asymptomatic, mild respiratory tract infection, to severe cases with acute respiratory distress syndrome, respiratory failure, and death. Reports on a dysregulated immune system in the severe cases call for a better characterization and understanding of the changes in the immune system.Methods: In order to dissect COVID-19-driven immune host responses, we performed RNA-seq of whole blood cell transcriptomes and granulocyte preparations from mild and severe COVID-19 patients and analyzed the data using a combination of conventional and data-driven co-expression analysis. Additionally, publicly available data was used to show the distinction from COVID-19 to other diseases. Reverse drug target prediction was used to identify known or novel drug candidates based on finding from data-driven findings.Results: Here, we profiled whole blood transcriptomes of 39 COVID-19 patients and 10 control donors enabling a data-driven stratification based on molecular phenotype. Neutrophil activation-associated signatures were prominently enriched in severe patient groups, which was corroborated in whole blood transcriptomes from an independent second cohort of 30 as well as in granulocyte samples from a third cohort of 16 COVID-19 patients (44 samples). Comparison of COVID-19 blood transcriptomes with those of a collection of over 3100 samples derived from 12 different viral infections, inflammatory diseases, and independent control samples revealed highly specific transcriptome signatures for COVID-19. Further, stratified transcriptomes predicted patient subgroup-specific drug candidates targeting the dysregulated systemic immune response of the host.Conclusions: Our study provides novel insights in the distinct molecular subgroups or phenotypes that are not simply explained by clinical parameters. We show that whole blood transcriptomes are extremely informative for COVID-19 since they capture granulocytes which are major drivers of disease severity.
16 July 2021 in EXCLI Journal
An increasing amount of attention has been geared towards understanding the privacy risks that arise from sharing genomic data of human origin. Most of these efforts have focused on issues in the context of genomic sequence data, but the popularity of techniques for collecting other types of genome-related data has prompted researchers to investigate privacy concerns in a broader genomic context. In this review, we give an overview of different types of genome-associated data, their individual ways of revealing sensitive information, the motivation to share them as well as established and upcoming methods to minimize information leakage. We further discuss the concise threats that are being posed, who is at risk, and how the risk level compares to potential benefits, all while addressing the topic in the context of modern technology, methodology, and information sharing culture. Additionally, we will discuss the current legal situation regarding the sharing of genomic data in a selection of countries, evaluating the scope of their applicability as well as their limitations. We will finalize this review by evaluating the development that is required in the scientific field in the near future in order to improve and develop privacy-preserving data sharing techniques for the genomic context.
5 March 2024 in Journal of Cheminformatics
Autoencoders are frequently used to embed molecules for training of downstream deep learning models. However, evaluation of the chemical information quality in the latent spaces is lacking and the model architectures are often arbitrarily chosen. Unoptimized architectures may not only negatively affect latent space quality but also increase energy consumption during training, making the models unsustainable. We conducted systematic experiments to better understand how the autoencoder architecture affects the reconstruction and latent space quality and how it can be optimized towards the encoding task as well as energy consumption. We can show that optimizing the architecture allows us to maintain the quality of a generic architecture but using 97% less data and reducing energy consumption by around 36%. We additionally observed that representing the molecules as SELFIES reduced the reconstruction performance compared to SMILES and that training with enumerated SMILES drastically improved latent space quality. Scientific Contribution: This work provides the first comprehensive systematic analysis of how choosing the autoencoder architecture affects the reconstruction performance of small molecules, the chemical information content of the latent space as well as the energy required for training. Demonstrated on the MOSES benchmarking dataset it provides first valuable insights into how autoencoders for the embedding of small molecules can be designed to optimize their utility and simultaneously become more sustainable, both in terms of energy consumption as well as the required amount of training data. All code, data and model checkpoints are made available on Zenodo (Oestreich et al. Small molecule autoencoders: architecture engineering to optimize latent space utility and sustainability. Zenodo, 2024). Furthermore, the top models can be found on GitHub with scripts to encode custom molecules: https://github.com/MarieOestreich/small-molecule-autoencoders. Graphical Abstract