Kramer, Frank

[["dc.bibliographiccitation.artnumber","1140"],["dc.bibliographiccitation.issue","6"],["dc.bibliographiccitation.journal","International Journal of Molecular Sciences"],["dc.bibliographiccitation.volume","18"],["dc.contributor.affiliation","Jo, Peter; \t\t \r\n\t\t Department of General-, Visceral-, and Pediatric Surgery, University Medical Center Goettingen, Robert-Koch-Str. 40, 37075 Goettingen, Germany, jo.peter@chirurgie-goettingen.de"],["dc.contributor.affiliation","Azizian, Azadeh; \t\t \r\n\t\t Department of General-, Visceral-, and Pediatric Surgery, University Medical Center Goettingen, Robert-Koch-Str. 40, 37075 Goettingen, Germany, azadeh.azizian@med.uni-goettingen.de"],["dc.contributor.affiliation","Salendo, Junius; \t\t \r\n\t\t Department of General-, Visceral-, and Pediatric Surgery, University Medical Center Goettingen, Robert-Koch-Str. 40, 37075 Goettingen, Germany, juniussalendo@gmail.com"],["dc.contributor.affiliation","Kramer, Frank; \t\t \r\n\t\t Department of Medical Statistics, University Medical Center Goettingen, Robert-Koch-Str. 40, 37075 Goettingen, Germany, frank.kramer@med.uni-goettingen.de"],["dc.contributor.affiliation","Bernhardt, Markus; \t\t \r\n\t\t Department of General-, Visceral-, and Pediatric Surgery, University Medical Center Goettingen, Robert-Koch-Str. 40, 37075 Goettingen, Germany, markus.bernhardt@med.uni-goettingen.de"],["dc.contributor.affiliation","Wolff, Hendrik; \t\t \r\n\t\t Department of Radiology, Nuclear Medicine and Radiotherapy, Radiology Munich, Burgstr. 7, 80333 Munich, Germany, drhawolff@googlemail.com"],["dc.contributor.affiliation","Gruber, Jens; \t\t \r\n\t\t German Primate Center, Medical RNA Biology, Kellnerweg 4, 37075 Goettingen, Germany, jgruber@dpz.eu"],["dc.contributor.affiliation","Grade, Marian; \t\t \r\n\t\t Department of General-, Visceral-, and Pediatric Surgery, University Medical Center Goettingen, Robert-Koch-Str. 40, 37075 Goettingen, Germany, marian.grade@med.uni-goettingen.de"],["dc.contributor.affiliation","Beißbarth, Tim; \t\t \r\n\t\t Department of Medical Statistics, University Medical Center Goettingen, Robert-Koch-Str. 40, 37075 Goettingen, Germany, tim.beissbarth@med.uni-goettingen.de"],["dc.contributor.affiliation","Ghadimi, B.; \t\t \r\n\t\t Department of General-, Visceral-, and Pediatric Surgery, University Medical Center Goettingen, Robert-Koch-Str. 40, 37075 Goettingen, Germany, mghadim@uni-goettingen.de"],["dc.contributor.affiliation","Gaedcke, Jochen; \t\t \r\n\t\t Department of General-, Visceral-, and Pediatric Surgery, University Medical Center Goettingen, Robert-Koch-Str. 40, 37075 Goettingen, Germany, jochen.gaedcke@med.uni-goettingen.de"],["dc.contributor.author","Jo, Peter"],["dc.contributor.author","Azizian, Azadeh"],["dc.contributor.author","Salendo, Junius"],["dc.contributor.author","Kramer, Frank"],["dc.contributor.author","Bernhardt, Markus"],["dc.contributor.author","Wolff, Hendrik Andreas"],["dc.contributor.author","Gruber, Jens"],["dc.contributor.author","Grade, Marian"],["dc.contributor.author","Beißbarth, Tim"],["dc.contributor.author","Ghadimi, Michael B."],["dc.contributor.author","Gaedcke, Jochen"],["dc.date.accessioned","2018-11-07T10:22:50Z"],["dc.date.available","2018-11-07T10:22:50Z"],["dc.date.issued","2017"],["dc.date.updated","2022-09-06T05:14:56Z"],["dc.description.abstract","Since the response to chemoradiotherapy in patients with locally advanced rectal cancer is heterogeneous, valid biomarkers are needed to monitor tumor response. Circulating microRNAs are promising candidates, however analyses of circulating microRNAs in rectal cancer are still rare. 111 patients with rectal cancer and 46 age-matched normal controls were enrolled. The expression levels of 30 microRNAs were analyzed in 17 pre-treatment patients' plasma samples. Differentially regulated microRNAs were validated in 94 independent patients. For 52 of the 94 patients a paired comparison between pre-treatment and post-treatment samples was performed. miR-17, miR-18b, miR-20a, miR-31, and miR-193a_3p, were significantly downregulated in pre-treatment plasma samples of patients with rectal cancer (p < 0.05). miR-29c, miR-30c, and miR-195 showed a trend of differential regulation. After validation, miR-31 and miR-30c were significantly deregulated by a decrease of expression. In 52 patients expression analyses of the 8 microRNAs in matched pre-treatment and post-treatment samples showed a significant decrease for all microRNAs (p < 0.05) after treatment. Expression levels of miR-31 and miR-30c could serve as valid biomarkers if validated in a prospective study. Plasma microRNA expression levels do not necessarily represent miRNA expression levels in tumor tissue. Also, expression levels of microRNAs change during multimodal therapy."],["dc.description.sponsorship","DFG (German Research Foundation)"],["dc.identifier.doi","10.3390/ijms18061140"],["dc.identifier.isi","000404581500040"],["dc.identifier.purl","https://resolver.sub.uni-goettingen.de/purl?gs-1/14793"],["dc.identifier.uri","https://resolver.sub.uni-goettingen.de/purl?gro-2/42349"],["dc.language.iso","en"],["dc.notes.intern","Merged from goescholar"],["dc.notes.status","zu prüfen"],["dc.notes.submitter","PUB_WoS_Import"],["dc.publisher","Mdpi Ag"],["dc.relation.eissn","1422-0067"],["dc.relation.issn","1422-0067"],["dc.rights","CC BY 4.0"],["dc.rights.uri","https://creativecommons.org/licenses/by/4.0"],["dc.title","Changes of Microrna Levels in Plasma of Patients with Rectal Cancer during Chemoradiotherapy"],["dc.type","journal_article"],["dc.type.internalPublication","yes"],["dc.type.peerReviewed","yes"],["dc.type.status","published"],["dc.type.subtype","original_ja"],["dc.type.version","published_version"],["dspace.entity.type","Publication"]]

[["dc.bibliographiccitation.artnumber","519"],["dc.bibliographiccitation.issue","11"],["dc.bibliographiccitation.journal","Genes"],["dc.bibliographiccitation.volume","9"],["dc.contributor.author","Hammoud, Zaynab"],["dc.contributor.author","Kramer, Frank"],["dc.date.accessioned","2019-02-19T13:26:29Z"],["dc.date.available","2019-02-19T13:26:29Z"],["dc.date.issued","2018"],["dc.description.abstract","The modelling of complex biological networks such as pathways has been a necessity for scientists over the last decades. The study of these networks also imposes a need to investigate different aspects of nodes or edges within the networks, or other biomedical knowledge related to it. Our aim is to provide a generic modelling framework to integrate multiple pathway types and further knowledge sources influencing these networks. This framework is defined by a multi-layered model allowing automatic network transformations and documentation. By providing a tool that generates this model, we aim to facilitate the data integration, boost the reproducibility and increase the interoperability between different sources and databases in the field of pathways. We present mully R package that allows the user to create, modify and visualize graphs with multi-layers. The package is implemented with features to specifically handle multilayered graphs."],["dc.description.sponsorship","Open-Access-Publikatinsfonds 2018"],["dc.identifier.doi","10.3390/genes9110519"],["dc.identifier.pmid","30360563"],["dc.identifier.purl","https://resolver.sub.uni-goettingen.de/purl?gs-1/15403"],["dc.identifier.uri","https://resolver.sub.uni-goettingen.de/purl?gro-2/57584"],["dc.language.iso","en"],["dc.notes.intern","Merged from goescholar"],["dc.notes.status","final"],["dc.publisher","MDPI"],["dc.relation.eissn","2073-4425"],["dc.rights","CC BY 4.0"],["dc.rights.uri","https://creativecommons.org/licenses/by/4.0"],["dc.title","mully: An R Package to Create, Modify and Visualize Multilayered Graphs"],["dc.type","journal_article"],["dc.type.internalPublication","yes"],["dc.type.version","published_version"],["dspace.entity.type","Publication"]]

[["dc.bibliographiccitation.issue","1"],["dc.bibliographiccitation.journal","BMC Systems Biology"],["dc.bibliographiccitation.volume","12"],["dc.contributor.author","Kramer, Frank"],["dc.contributor.author","Just, Steffen"],["dc.contributor.author","Zeller, Tanja"],["dc.date.accessioned","2020-12-10T18:38:58Z"],["dc.date.available","2020-12-10T18:38:58Z"],["dc.date.issued","2018"],["dc.identifier.doi","10.1186/s12918-018-0579-5"],["dc.identifier.eissn","1752-0509"],["dc.identifier.purl","https://resolver.sub.uni-goettingen.de/purl?gs-1/15493"],["dc.identifier.uri","https://resolver.sub.uni-goettingen.de/purl?gro-2/77497"],["dc.language.iso","en"],["dc.notes.intern","DOI Import GROB-354"],["dc.notes.intern","Merged from goescholar"],["dc.rights","CC BY 4.0"],["dc.rights.uri","https://creativecommons.org/licenses/by/4.0"],["dc.title","New perspectives: systems medicine in cardiovascular disease"],["dc.type","journal_article"],["dc.type.internalPublication","yes"],["dc.type.version","published_version"],["dspace.entity.type","Publication"]]

[["dc.bibliographiccitation.artnumber","568"],["dc.bibliographiccitation.issue","4"],["dc.bibliographiccitation.journal","International Journal of Molecular Sciences"],["dc.bibliographiccitation.volume","17"],["dc.contributor.author","Azizian, Azadeh"],["dc.contributor.author","Epping, Ingo"],["dc.contributor.author","Kramer, Frank"],["dc.contributor.author","Jo, Peter"],["dc.contributor.author","Bernhardt, Markus"],["dc.contributor.author","Kitz, Julia"],["dc.contributor.author","Salinas, Gabriela"],["dc.contributor.author","Wolff, Hendrik Andreas"],["dc.contributor.author","Grade, Marian"],["dc.contributor.author","Beissbarath, Tim"],["dc.contributor.author","Ghadimi, B. Michael"],["dc.contributor.author","Gaedcke, Jochen"],["dc.date.accessioned","2018-11-07T10:16:01Z"],["dc.date.available","2018-11-07T10:16:01Z"],["dc.date.issued","2016"],["dc.description.abstract","Background: Patients with locally advanced rectal cancer are treated with preoperative chemoradiotherapy followed by surgical resection. Despite similar clinical parameters (uT2-3, uN+) and standard therapy, patients' prognoses differ widely. A possible prediction of prognosis through microRNAs as biomarkers out of treatment-naive biopsies would allow individualized therapy options. Methods: Microarray analysis of 45 microdissected preoperative biopsies from patients with rectal cancer was performed to identify potential microRNAs to predict overall survival, disease-free survival, cancer-specific survival, distant-metastasis-free survival, tumor regression grade, or nodal stage. Quantitative real-time polymerase chain reaction (qPCR) was performed on an independent set of 147 rectal cancer patients to validate relevant miRNAs. Results: In the microarray screen, 14 microRNAs were significantly correlated to overall survival. Five microRNAs were included from previous work. Finally, 19 miRNAs were evaluated by qPCR. miR-515-5p, miR-573, miR-579 and miR-802 demonstrated significant correlation with overall survival and cancer-specific survival (p<0.05). miR-573 was also significantly correlated with the tumor regression grade after preoperative chemoradiotherapy. miR-133b showed a significant correlation with distant-metastasis-free survival. miR-146b expression levels showed a significant correlation with nodal stage. Conclusion: Specific microRNAs can be used as biomarkers to predict prognosis of patients with rectal cancer and possibly stratify patients' therapy if validated in a prospective study."],["dc.identifier.doi","10.3390/ijms17040568"],["dc.identifier.isi","000374585300147"],["dc.identifier.pmid","27092493"],["dc.identifier.purl","https://resolver.sub.uni-goettingen.de/purl?gs-1/13222"],["dc.identifier.uri","https://resolver.sub.uni-goettingen.de/purl?gro-2/40949"],["dc.notes.intern","Merged from goescholar"],["dc.notes.status","zu prüfen"],["dc.notes.submitter","Najko"],["dc.publisher","Mdpi Ag"],["dc.relation.issn","1422-0067"],["dc.rights","CC BY 4.0"],["dc.rights.uri","https://creativecommons.org/licenses/by/4.0"],["dc.title","Prognostic Value of MicroRNAs in Preoperative Treated Rectal Cancer"],["dc.type","journal_article"],["dc.type.internalPublication","yes"],["dc.type.peerReviewed","yes"],["dc.type.status","published"],["dc.type.version","published_version"],["dspace.entity.type","Publication"]]

[["dc.bibliographiccitation.artnumber","13"],["dc.bibliographiccitation.journal","Epigenetics & Chromatin"],["dc.bibliographiccitation.volume","5"],["dc.contributor.author","Prenzel, Tanja"],["dc.contributor.author","Kramer, Frank"],["dc.contributor.author","Bedi, Upasana"],["dc.contributor.author","Nagarajan, Sankari"],["dc.contributor.author","Beißbarth, Tim"],["dc.contributor.author","Johnsen, Steven Arthur"],["dc.date.accessioned","2018-11-07T09:07:07Z"],["dc.date.available","2018-11-07T09:07:07Z"],["dc.date.issued","2012"],["dc.description.abstract","Background: In conjunction with posttranslational chromatin modifications, proper arrangement of higher order chromatin structure appears to be important for controlling transcription in the nucleus. Recent genome-wide studies have shown that the Estrogen Receptor-alpha (ER alpha), encoded by the ESR1 gene, nucleates tissue-specific long-range chromosomal interactions in collaboration with the cohesin complex. Furthermore, the Mediator complex not only regulates ERa activity, but also interacts with the cohesin complex to facilitate long-range chromosomal interactions. However, whether the cohesin and Mediator complexes function together to contribute to estrogen-regulated gene transcription remains unknown. Results: In this study we show that depletion of the cohesin subunit SMC3 or the Mediator subunit MED12 significantly impairs the ER alpha-regulated transcriptome. Surprisingly, SMC3 depletion appears to elicit this effect indirectly by rapidly decreasing ESR1 transcription and ER alpha protein levels. Moreover, we provide evidence that both SMC3 and MED12 colocalize on the ESR1 gene and are mutually required for their own occupancy as well as for RNAPII occupancy across the ESR1 gene. Finally, we show that extended proteasome inhibition decreases the mRNA expression of cohesin subunits which accompanies a decrease in ESR1 mRNA and ERa protein levels as well as estrogen-regulated transcription. Conclusions: These results identify the ESR1 gene as a cohesin/Mediator-dependent gene and indicate that this regulation may potentially be exploited for the treatment of estrogen-dependent breast cancer."],["dc.identifier.doi","10.1186/1756-8935-5-13"],["dc.identifier.isi","000310834100001"],["dc.identifier.pmid","22913342"],["dc.identifier.purl","https://resolver.sub.uni-goettingen.de/purl?gs-1/12877"],["dc.identifier.uri","https://resolver.sub.uni-goettingen.de/purl?gro-2/25716"],["dc.notes.intern","Merged from goescholar"],["dc.notes.status","zu prüfen"],["dc.notes.submitter","Najko"],["dc.publisher","Biomed Central Ltd"],["dc.relation.issn","1756-8935"],["dc.rights","CC BY 2.0"],["dc.rights.uri","https://creativecommons.org/licenses/by/2.0"],["dc.title","Cohesin is required for expression of the estrogen receptor-alpha (ESR1) gene"],["dc.type","journal_article"],["dc.type.internalPublication","yes"],["dc.type.peerReviewed","yes"],["dc.type.status","published"],["dc.type.version","published_version"],["dspace.entity.type","Publication"]]

[["dc.bibliographiccitation.firstpage","1184"],["dc.bibliographiccitation.issue","4"],["dc.bibliographiccitation.journal","International Journal of Radiation Oncology*Biology*Physics"],["dc.bibliographiccitation.lastpage","1192"],["dc.bibliographiccitation.volume","78"],["dc.contributor.author","Spitzner, Melanie"],["dc.contributor.author","Emons, Georg"],["dc.contributor.author","Kramer, Frank"],["dc.contributor.author","Gaedcke, Jochen"],["dc.contributor.author","Rave-Fränk, Margret"],["dc.contributor.author","Scharf, Jens-Gerd"],["dc.contributor.author","Burfeind, Peter"],["dc.contributor.author","Becker, Heinz"],["dc.contributor.author","Beißbarth, Tim"],["dc.contributor.author","Ghadimi, Michael B."],["dc.contributor.author","Ried, Thomas"],["dc.contributor.author","Grade, Marian"],["dc.date.accessioned","2018-11-07T08:36:51Z"],["dc.date.available","2018-11-07T08:36:51Z"],["dc.date.issued","2010"],["dc.description.abstract","Purpose: The standard treatment of patients with locally advanced rectal cancers comprises preoperative 5-fluorouracil based chemoradiotherapy followed by standardized surgery. However, tumor response to multimodal treatment has varied greatly, ranging from complete resistance to complete pathologic regression. The prediction of the response is, therefore, an important clinical need. Methods and Materials: To establish in vitro models for studying the molecular basis of this heterogeneous tumor response, we exposed 12 colorectal cancer cell lines to 3 mu M of 5-fluorouracil and 2 Gy of radiation. The differences in treatment sensitivity were then correlated with the pretherapeutic gene expression profiles of these cell lines. Results: We observed a heterogeneous response, with surviving fractions ranging from 0.28 to 0.81, closely recapitulating clinical reality. Using a linear model analysis, we identified 4,796 features whose expression levels correlated significantly with the sensitivity to chemoradiotherapy (Q < .05), including many genes involved in the mitogen-activated protein kinase signaling pathway or cell cycle genes. These data have suggested a potential relevance of the insulin and Wnt signaling pathways for treatment response, and we identified STAT3, RASSF1, DOK3, and ERBB2 as potential therapeutic targets. The microarray measurements were independently validated for a subset of these genes using real-time polymerase chain reactions. Conclusion: We are the first to report a gene expression signature for the in vitro chemoradiosensitivity of colorectal cancer cells. We anticipate that this analysis will unveil molecular biomarkers predictive of the response of rectal cancers to chemoradiotherapy and enable the identification of genes that could serve as targets to sensitize a priori resistant primary tumors. (C) 2010 Elsevier Inc."],["dc.description.sponsorship","Deutsche Forschungsgemeinschaft [KFO 179]"],["dc.identifier.doi","10.1016/j.ijrobp.2010.06.023"],["dc.identifier.isi","000283963100030"],["dc.identifier.pmid","20970032"],["dc.identifier.purl","https://resolver.sub.uni-goettingen.de/purl?gs-1/6106"],["dc.identifier.uri","https://resolver.sub.uni-goettingen.de/purl?gro-2/18405"],["dc.notes.intern","Merged from goescholar"],["dc.notes.status","zu prüfen"],["dc.notes.submitter","Najko"],["dc.publisher","Elsevier Science Inc"],["dc.relation.issn","0360-3016"],["dc.rights","Goescholar"],["dc.rights.uri","https://goescholar.uni-goettingen.de/licenses"],["dc.title","A GENE EXPRESSION SIGNATURE FOR CHEMORADIOSENSITIVITY OF COLORECTAL CANCER CELLS"],["dc.type","journal_article"],["dc.type.internalPublication","yes"],["dc.type.peerReviewed","yes"],["dc.type.status","published"],["dspace.entity.type","Publication"]]

[["dc.bibliographiccitation.artnumber","334"],["dc.bibliographiccitation.journal","BMC Bioinformatics"],["dc.bibliographiccitation.volume","16"],["dc.contributor.author","Bayerlová, Michaela"],["dc.contributor.author","Jung, Klaus"],["dc.contributor.author","Kramer, Frank"],["dc.contributor.author","Klemm, Florian"],["dc.contributor.author","Bleckmann, Annalen"],["dc.contributor.author","Beißbarth, Tim"],["dc.date.accessioned","2018-11-07T09:50:02Z"],["dc.date.available","2018-11-07T09:50:02Z"],["dc.date.issued","2015"],["dc.description.abstract","Background: Enrichment analysis is a popular approach to identify pathways or sets of genes which are significantly enriched in the context of differentially expressed genes. The traditional gene set enrichment approach considers a pathway as a simple gene list disregarding any knowledge of gene or protein interactions. In contrast, the new group of so called pathway topology-based methods integrates the topological structure of a pathway into the analysis. Methods: We comparatively investigated gene set and pathway topology-based enrichment approaches, considering three gene set and four topological methods. These methods were compared in two extensive simulation studies and on a benchmark of 36 real datasets, providing the same pathway input data for all methods. Results: In the benchmark data analysis both types of methods showed a comparable ability to detect enriched pathways. The first simulation study was conducted with KEGG pathways, which showed considerable gene overlaps between each other. In this study with original KEGG pathways, none of the topology-based methods outperformed the gene set approach. Therefore, a second simulation study was performed on non-overlapping pathways created by unique gene IDs. Here, methods accounting for pathway topology reached higher accuracy than the gene set methods, however their sensitivity was lower. Conclusions: We conducted one of the first comprehensive comparative works on evaluating gene set against pathway topology-based enrichment methods. The topological methods showed better performance in the simulation scenarios with non-overlapping pathways, however, they were not conclusively better in the other scenarios. This suggests that simple gene set approach might be sufficient to detect an enriched pathway under realistic circumstances. Nevertheless, more extensive studies and further benchmark data are needed to systematically evaluate these methods and to assess what gain and cost pathway topology information introduces into enrichment analysis. Both types of methods for enrichment analysis require further improvements in order to deal with the problem of pathway overlaps."],["dc.description.sponsorship","Open-Access Publikationsfonds 2015"],["dc.identifier.doi","10.1186/s12859-015-0751-5"],["dc.identifier.isi","000363615900001"],["dc.identifier.pmid","26489510"],["dc.identifier.purl","https://resolver.sub.uni-goettingen.de/purl?gs-1/12346"],["dc.identifier.uri","https://resolver.sub.uni-goettingen.de/purl?gro-2/35629"],["dc.notes.intern","Merged from goescholar"],["dc.notes.status","zu prüfen"],["dc.notes.submitter","Najko"],["dc.publisher","Biomed Central Ltd"],["dc.relation.issn","1471-2105"],["dc.rights","CC BY 4.0"],["dc.rights.uri","https://creativecommons.org/licenses/by/4.0"],["dc.title","Comparative study on gene set and pathway topology-based enrichment methods"],["dc.type","journal_article"],["dc.type.internalPublication","yes"],["dc.type.peerReviewed","yes"],["dc.type.status","published"],["dc.type.version","published_version"],["dspace.entity.type","Publication"]]

[["dc.bibliographiccitation.artnumber","42"],["dc.bibliographiccitation.issue","1"],["dc.bibliographiccitation.journal","Genome Medicine"],["dc.bibliographiccitation.volume","13"],["dc.contributor.author","Chereda, Hryhorii"],["dc.contributor.author","Bleckmann, Annalen"],["dc.contributor.author","Menck, Kerstin"],["dc.contributor.author","Perera-Bel, Júlia"],["dc.contributor.author","Stegmaier, Philip"],["dc.contributor.author","Auer, Florian"],["dc.contributor.author","Kramer, Frank"],["dc.contributor.author","Leha, Andreas"],["dc.contributor.author","Beißbarth, Tim"],["dc.date.accessioned","2021-04-14T08:28:07Z"],["dc.date.accessioned","2022-08-18T12:39:45Z"],["dc.date.available","2021-04-14T08:28:07Z"],["dc.date.available","2022-08-18T12:39:45Z"],["dc.date.issued","2021-03-11"],["dc.date.updated","2022-07-29T12:18:05Z"],["dc.description.abstract","Abstract\r\n \r\n Background\r\n Contemporary deep learning approaches show cutting-edge performance in a variety of complex prediction tasks. Nonetheless, the application of deep learning in healthcare remains limited since deep learning methods are often considered as non-interpretable black-box models. However, the machine learning community made recent elaborations on interpretability methods explaining data point-specific decisions of deep learning techniques. We believe that such explanations can assist the need in personalized precision medicine decisions via explaining patient-specific predictions.\r\n \r\n \r\n Methods\r\n Layer-wise Relevance Propagation (LRP) is a technique to explain decisions of deep learning methods. It is widely used to interpret Convolutional Neural Networks (CNNs) applied on image data. Recently, CNNs started to extend towards non-Euclidean domains like graphs. Molecular networks are commonly represented as graphs detailing interactions between molecules. Gene expression data can be assigned to the vertices of these graphs. In other words, gene expression data can be structured by utilizing molecular network information as prior knowledge. Graph-CNNs can be applied to structured gene expression data, for example, to predict metastatic events in breast cancer. Therefore, there is a need for explanations showing which part of a molecular network is relevant for predicting an event, e.g., distant metastasis in cancer, for each individual patient.\r\n \r\n \r\n Results\r\n We extended the procedure of LRP to make it available for Graph-CNN and tested its applicability on a large breast cancer dataset. We present Graph Layer-wise Relevance Propagation (GLRP) as a new method to explain the decisions made by Graph-CNNs. We demonstrate a sanity check of the developed GLRP on a hand-written digits dataset and then apply the method on gene expression data. We show that GLRP provides patient-specific molecular subnetworks that largely agree with clinical knowledge and identify common as well as novel, and potentially druggable, drivers of tumor progression.\r\n \r\n \r\n Conclusions\r\n The developed method could be potentially highly useful on interpreting classification results in the context of different omics data and prior knowledge molecular networks on the individual patient level, as for example in precision medicine approaches or a molecular tumor board."],["dc.description.sponsorship","Open-Access-Publikationsfonds 2021"],["dc.identifier.citation","Genome Medicine. 2021 Mar 11;13(1):42"],["dc.identifier.doi","10.1186/s13073-021-00845-7"],["dc.identifier.purl","https://resolver.sub.uni-goettingen.de/purl?gs-1/17744"],["dc.identifier.uri","https://resolver.sub.uni-goettingen.de/purl?gro-2/82506"],["dc.identifier.uri","https://resolver.sub.uni-goettingen.de/purl?gro-2/112973"],["dc.language.iso","en"],["dc.notes.intern","DOI Import GROB-399"],["dc.notes.intern","Merged from goescholar"],["dc.publisher","BioMed Central"],["dc.relation.eissn","1756-994X"],["dc.rights","CC BY 4.0"],["dc.rights.holder","The Author(s)"],["dc.rights.uri","https://creativecommons.org/licenses/by/4.0"],["dc.subject","Gene expression data"],["dc.subject","Explainable AI"],["dc.subject","Personalized medicine"],["dc.subject","Precision medicine"],["dc.subject","Classification of cancer"],["dc.subject","Deep learning"],["dc.subject","Prior knowledge"],["dc.subject","Molecular networks"],["dc.title","Explaining decisions of graph convolutional neural networks: patient-specific molecular subnetworks responsible for metastasis prediction in breast cancer"],["dc.type","journal_article"],["dc.type.internalPublication","yes"],["dc.type.version","published_version"],["dspace.entity.type","Publication"]]

[["dc.bibliographiccitation.artnumber","e0144014"],["dc.bibliographiccitation.issue","12"],["dc.bibliographiccitation.journal","PLoS ONE"],["dc.bibliographiccitation.volume","10"],["dc.contributor.author","Bayerlová, Michaela"],["dc.contributor.author","Klemm, Florian"],["dc.contributor.author","Kramer, Frank"],["dc.contributor.author","Pukrop, Tobias"],["dc.contributor.author","Beißbarth, Tim"],["dc.contributor.author","Bleckmann, Annalen"],["dc.date.accessioned","2018-11-07T09:47:50Z"],["dc.date.available","2018-11-07T09:47:50Z"],["dc.date.issued","2015"],["dc.description.abstract","Introduction WNT signaling is a complex process comprising multiple pathways: the canonical beta-catenin- dependent pathway and several alternative non-canonical pathways that act in a beta-catenin- independent manner. Representing these intricate signaling mechanisms through bioinformatic approaches is challenging. Nevertheless, a simplified but reliable bioinformatic WNT pathway model is needed, which can be further utilized to decipher specific WNT activation states within e.g. high-throughput data. Results In order to build such a model, we collected, parsed, and curated available WNT signaling knowledge from different pathway databases. The data were assembled to construct computationally suitable models of different WNT signaling cascades in the form of directed signaling graphs. This resulted in four networks representing canonical WNT signaling, non-canonical WNT signaling, the inhibition of canonical WNT signaling and the regulation of WNT signaling pathways, respectively. Furthermore, these networks were integrated with microarray and RNA sequencing data to gain deeper insight into the underlying biology of gene expression differences between MCF-7 and MDA-MB-231 breast cancer cell lines, representing weakly and highly invasive breast carcinomas, respectively. Differential genes up-regulated in the MDA-MB-231 compared to the MCF-7 cell line were found to display enrichment in the gene set originating from the non-canonical network. Moreover, we identified and validated differentially regulated modules representing canonical and non-canonical WNT pathway components specific for the aggressive basal-like breast cancer subtype. Conclusions In conclusion, we demonstrated that these newly constructed WNT networks reliably reflect distinct WNT signaling processes. Using transcriptomic data, we shaped these networks into comprehensive modules of the genes implicated in the aggressive basal-like breast cancer subtype and demonstrated that non-canonical WNT signaling is important in this context. The topology of these networks can be further refined in the future by integration with complementary data such as protein-protein interactions, in order to gain greater insight into signaling processes."],["dc.description.sponsorship","Open-Access-Publikationsfonds 2015"],["dc.identifier.doi","10.1371/journal.pone.0144014"],["dc.identifier.isi","000366040000042"],["dc.identifier.pmid","26632845"],["dc.identifier.purl","https://resolver.sub.uni-goettingen.de/purl?gs-1/12616"],["dc.identifier.uri","https://resolver.sub.uni-goettingen.de/purl?gro-2/35182"],["dc.notes.intern","Merged from goescholar"],["dc.notes.status","zu prüfen"],["dc.notes.submitter","Najko"],["dc.publisher","Public Library Science"],["dc.relation.issn","1932-6203"],["dc.rights","CC BY 4.0"],["dc.rights.uri","https://creativecommons.org/licenses/by/4.0"],["dc.title","Newly Constructed Network Models of Different WNT Signaling Cascades Applied to Breast Cancer Expression Data"],["dc.type","journal_article"],["dc.type.internalPublication","yes"],["dc.type.peerReviewed","yes"],["dc.type.status","published"],["dc.type.version","published_version"],["dspace.entity.type","Publication"]]

[["dc.bibliographiccitation.firstpage","85"],["dc.bibliographiccitation.issue","1"],["dc.bibliographiccitation.journal","Biology"],["dc.bibliographiccitation.lastpage","100"],["dc.bibliographiccitation.volume","3"],["dc.contributor.author","Kramer, Frank"],["dc.contributor.author","Bayerlová, Michaela"],["dc.contributor.author","Beißbarth, Tim"],["dc.date.accessioned","2019-07-09T11:41:04Z"],["dc.date.available","2019-07-09T11:41:04Z"],["dc.date.issued","2014"],["dc.description.abstract","Putting new findings into the context of available literature knowledge is one approach to deal with the surge of high-throughput data results. Furthermore, prior knowledge can increase the performance and stability of bioinformatic algorithms, for example, methods for network reconstruction. In this review, we examine software packages for the statistical computing framework R, which enable the integration of pathway data for further bioinformatic analyses. Different approaches to integrate and visualize pathway data are identified and packages are stratified concerning their features according to a number of different aspects: data import strategies, the extent of available data, dependencies on external tools, integration with further analysis steps and visualization options are considered. A total of 12 packages integrating pathway data are reviewed in this manuscript. These are supplemented by five R-specific packages for visualization and six connector packages, which provide access to external tools."],["dc.identifier.doi","10.3390/biology3010085"],["dc.identifier.pmid","24833336"],["dc.identifier.purl","https://resolver.sub.uni-goettingen.de/purl?gs-1/11669"],["dc.identifier.uri","https://resolver.sub.uni-goettingen.de/purl?gro-2/58353"],["dc.language.iso","en"],["dc.notes.intern","Merged from goescholar"],["dc.relation.issn","2079-7737"],["dc.rights","CC BY 3.0"],["dc.rights.uri","https://creativecommons.org/licenses/by/3.0"],["dc.title","R-based software for the integration of pathway data into bioinformatic algorithms."],["dc.type","journal_article"],["dc.type.internalPublication","yes"],["dc.type.version","published_version"],["dspace.entity.type","Publication"]]