Biology subjects

Trautwein, C.

Publications and source records attributed to Trautwein, C..

3 recordsLinked to original sources

bioRxiv2019

Pan-cancer image-based detection of clinically actionable genetic alterations

Precision treatment of cancer relies on genetic alterations which are diagnosed by molecular biology assays.1 These tests can be a bottleneck in oncology workflows because of high turnaround time, tissue usage and costs.2 Here, we show that deep learning can predict point mutations, molecular tumor subtypes and immune-related gene expression signatures3,4 directly from routine histological images of tumor tissue. We developed and systematically optimized a one-stop-shop workflow and applied it to more than 4000 patients with breast5, colon and rectal6, head and neck7, lung8,9, pancreatic10, prostate11 cancer, melanoma12 and gastric13 cancer. Together, our findings show that a single deep learning algorithm can predict clinically actionable alterations from routine histology data. Our method can be implemented on mobile hardware14, potentially enabling point-of-care diagnostics for personalized cancer treatment in individual patients.

genetics↗

bioRxiv2019

DJ-1 (Park7) affects the gut microbiome, metabolites, and development of Innate Lymphoid Cells (ILCs)

The proper communication between gut and brain is pivotal for maintenance of health and dysregulation of the gut-brain axis can lead to several clinical disorders. Also, in Parkinsons disease (PD) 85% of all patients experienced constipation long before showing any signs of motor phenotypes. For differential diagnosis and when it comes to preventive treatment there is an urgent need for the identification of biomarkers indicating early disease stages long before the disease phenotype manifests. DJ-1 is a chaperon protein involved in the protection against PD and genetic mutations in this protein have been shown to cause familial PD. However, how the deficiency of DJ-1 modifies the PD risk remains incompletely understood. In the present study we provide evidence that DJ-1 is implicated in shaping the gut microbiome including their metabolite production or innate immune cells (ILCs) development. We revealed that in 4 months old mice genetic deficiency of DJ-1 leads to significantly decrease in several bacterial genera and significantly increase in two specific genera, namely Alistipes and Rikenella. DJ-1 deficient mice have a higher production of calprotectin/MCP-1 inflammatory protein - a known protein involved in colonic inflammation - and significantly higher expression of glial fibrillary acidic protein (GFAP) than control littermates. Expression of a-Synuclein, a key protein in Lewy bodies, in the colon was not significantly different between genotypes. Metabolic profiles of feces extracts analysed by H1-NMR spectroscopy showed increased short chain fatty acids (SCFAs) and decreased amino acid levels, suggesting a general switch from protein towards fibre degrading strains in DJ-1 deficient mice. We observed that Malonate - which is known to influence the immune system - has significantly higher concentrations in DJ-1 deficient mice. Moreover, DJ-1 deficient mice have high levels of the phenol derivate 3-(3-Hydroxyphenyl) propanoic acid (3-HPPA) which is a breakdown product of aromatic substrates like tyrosine, phenylalanine and polyphenols. DJ-1 deficient mice also showed significantly reduced percentage of ILCs. Thus, our data suggests that absence of DJ-1 leads to increase in gut inflammatory bacteria composition, deregulated metabolites and dysregulated innate immunity which could be a key factor in the initiation of PD disease in the gut, and potentially also in brain during disease progression.

immunology↗

bioRxiv2019

Deep learning detects virus presence in cancer histology

Oncogenic viruses like human papilloma virus (HPV) or Epstein Barr virus (EBV) are a major cause of human cancer. Viral oncogenesis has a direct impact on treatment decisions because virus-associated tumors can demand a lower intensity of chemotherapy and radiation or can be more susceptible to immune check-point inhibition. However, molecular tests for HPV and EBV are not ubiquitously available.\n\nWe hypothesized that the histopathological features of virus-driven and non-virus driven cancers are sufficiently different to be detectable by artificial intelligence (AI) through deep learning-based analysis of images from routine hematoxylin and eosin (HE) stained slides. We show that deep transfer learning can predict presence of HPV in head and neck cancer with a patient-level 3-fold cross validated area-under-the-curve (AUC) of 0.89 [0.82; 0.94]. The same workflow was used for Epstein-Barr virus (EBV) driven gastric cancer achieving a cross-validated AUC of 0.80 [0.70; 0.92] and a similar performance in external validation sets. Reverse-engineering our deep neural networks, we show that the key morphological features can be made understandable to humans.\n\nThis workflow could enable a fast and low-cost method to identify virus-induced cancer in clinical trials or clinical routine. At the same time, our approach for feature visualization allows pathologists to look into the black box of deep learning, enabling them to check the plausibility of computer-based image classification.

cancer biology↗