Deriving chemosensitivity from cell lines: Forensic bioinformatics and reproducible research in high-throughput biology

摘要: High-throughput biological assays such as microarrays let us ask very detailed questions about how diseases operate, and promise to personalize therapy. Data processing, however, is often not described well enough allow for exact reproduction of the results, leading exercises in “forensic bioinformatics” where aspects raw data reported results are used infer what methods must have been employed. Unfortunately, poor documentation can shift from an inconvenience active danger when it obscures just but errors. In this report, we examine several related papers purporting use microarray-based signatures drug sensitivity derived cell lines predict patient response. Patients clinical trials currently being allocated treatment arms on basis these results. However, show five case studies that incorporate simple errors may be putting patients at risk. One theme emerges most common (e.g., row or column offsets); conversely, our experience common. We then discuss steps taking avoid own investigations.