Refine
Year of publication
- 2023 (4) (remove)
Language
- English (4)
Has Fulltext
- yes (4)
Is part of the Bibliography
- no (4)
Institute
- Medizin (4) (remove)
Investigators in the cognitive neurosciences have turned to Big Data to address persistent replication and reliability issues by increasing sample sizes, statistical power, and representativeness of data. While there is tremendous potential to advance science through open data sharing, these efforts unveil a host of new questions about how to integrate data arising from distinct sources and instruments. We focus on the most frequently assessed area of cognition - memory testing - and demonstrate a process for reliable data harmonization across three common measures. We aggregated raw data from 53 studies from around the world which measured at least one of three distinct verbal learning tasks, totaling N = 10,505 healthy and brain-injured individuals. A mega analysis was conducted using empirical bayes harmonization to isolate and remove site effects, followed by linear models which adjusted for common covariates. After corrections, a continuous item response theory (IRT) model estimated each individual subject’s latent verbal learning ability while accounting for item difficulties. Harmonization significantly reduced inter-site variance by 37% while preserving covariate effects. The effects of age, sex, and education on scores were found to be highly consistent across memory tests. IRT methods for equating scores across AVLTs agreed with held-out data of dually-administered tests, and these tools are made available for free online. This work demonstrates that large-scale data sharing and harmonization initiatives can offer opportunities to address reproducibility and integration challenges across the behavioral sciences.
Reliable, easy-to-handle phenotypic screening platforms are needed for the identification of anti-SARS-CoV-2 compounds. Here, we present caspase 3/7 activity as a readout for monitoring the replication of SARS-CoV-2 isolates from different variants, including a remdesivir-resistant strain, and of other coronaviruses in numerous cell culture models, independently of cytopathogenic effect formation. Compared to other models, the Caco-2 subline Caco-2-F03 displayed superior performance. It possesses a stable SARS-CoV-2 susceptibility phenotype and does not produce false-positive hits due to drug-induced phospholipidosis. A proof-of-concept screen of 1,796 kinase inhibitors identified known and novel antiviral drug candidates including inhibitors of phosphoglycerate dehydrogenase (PHGDH), CDC like kinase 1 (CLK-1), and colony stimulating factor 1 receptor (CSF1R). The activity of the PHGDH inhibitor NCT-503 was further increased in combination with the hexokinase II (HK2) inhibitor 2-deoxy-D-glucose, which is in clinical development for COVID-19. In conclusion, caspase 3/7 activity detection in SARS-CoV-2-infected Caco-2-F03 cells provides a simple phenotypic high-throughput screening platform for SARS-CoV-2 drug candidates that reduces false-positive hits.
In the life sciences, there is an ongoing discussion about a perceived ‘reproducibility crisis’. However, it remains unclear to which extent the perceived lack of reproducibility is the consequence of issues that can be tackled and to which extent it may be the consequence of unrealistic expectations of the technical level of reproducibility. Large-scale, multi-institutional experimental replication studies are very cost- and time-intensive. This Perspective suggests an alternative, complementary approach: meta-research using sociological and philosophical methodologies to examine researcher trust in data. An improved understanding of the criteria used by researchers to judge data reliability will provide crucial, initial evidence on the actual scale of the reproducibility crisis and on measures to tackle it.