Dataset Information

Comprehensive evaluation of composite gene features in cancer outcome prediction.

ABSTRACT: Owing to the heterogeneous and continuously evolving nature of cancers, classifiers based on the expression of individual genes usually do not result in robust prediction of cancer outcome. As an alternative, composite gene features that combine functionally related genes have been proposed. It is expected that such features can be more robust and reproducible since they can capture the alterations in relevant biological processes as a whole and may be less sensitive to fluctuations in the expression of individual genes. Various algorithms have been developed for the identification of composite features and inference of composite gene feature activity, which all claim to improve the prediction accuracy. However, because of the limitations of test datasets incorporated by each individual study and inconsistent test procedures, the results of these studies are sometimes conflicting and unproducible. For this reason, it is difficult to have a comprehensive understanding of the prediction performance of composite gene features, particularly across different cancers, cancer subtypes, and cohorts. In this study, we implement various algorithms for the identification of composite gene features and their utilization in cancer outcome prediction, and perform extensive comparison and evaluation using seven microarray datasets covering two cancer types and three different phenotypes. Our results show that, while some algorithms outperform others for certain classification tasks, no single algorithm consistently outperforms other algorithms and individual gene features.

SUBMITTER: Hou D

PROVIDER: S-EPMC4345828 | biostudies-literature | 2014

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Comprehensive evaluation of composite gene features in cancer outcome prediction.

Hou Dezhi D Koyutürk Mehmet M

Cancer informatics 20140101 Suppl 3

Owing to the heterogeneous and continuously evolving nature of cancers, classifiers based on the expression of individual genes usually do not result in robust prediction of cancer outcome. As an alternative, composite gene features that combine functionally related genes have been proposed. It is expected that such features can be more robust and reproducible since they can capture the alterations in relevant biological processes as a whole and may be less sensitive to fluctuations in the expre ...[more]

PMID: 25780335

Similar Datasets

Project description:ObjectiveThe aim of this study is to define and assess Ideal Outcome in the national or multicenter registries of North America, Germany, the Netherlands, and Sweden.BackgroundAssessing outcomes after pancreatoduodenectomy among centers and countries requires a broad evaluation that cannot be captured by a single parameter. Previously, 2 composite outcome measures (textbook outcome and optimal pancreatic surgery) for pancreatoduodenectomy have been described from Europe and the United States. These composites were harmonized into ideal outcome (IO).MethodsThis analysis is a transatlantic retrospective study (2018-2020) of patients after pancreatoduodenectomy within the registries from North America, Germany, The Netherlands, and Sweden. After 3 consensus meetings, IO for pancreatoduodenectomy was defined as the absence of all 6 parameters: (1) in-hospital mortality, (2) severe complications-Clavien-Dindo ≥3, (3) postoperative pancreatic fistula-International Study Group of Pancreatic Surgery (ISGPS) grade B/C, (4) reoperation, (5) hospital stay >75th percentile, and (6) readmission. Outcomes were evaluated using relative largest difference (RLD) and absolute largest difference (ALD), and multivariate regression models.ResultsOverall, 21,036 patients after pancreatoduodenectomy were included, of whom 11,194 (54%) reached IO. The rate of IO varied between 55% in North America, 53% in Germany, 52% in The Netherlands, and 54% in Sweden (RLD: 1.1, ALD: 3%, P <0.001). Individual components varied with an ALD of 2% length of stay, 4% for in-hospital mortality, 12% severe complications, 10% postoperative pancreatic fistula, 11% reoperation, and 9% readmission. Age, sex, absence of chronic obstructive pulmonary disease, body mass index, performance status, American Society of Anesthesiologists (ASA) score, biliary drainage, absence of vascular resection, and histologic diagnosis were associated with IO. In the subgroup of patients with pancreatic adenocarcinoma, country, and neoadjuvant chemotherapy also was associated with improved IO.ConclusionsThe newly developed composite outcome measure "Ideal Outcome" can be used for auditing and comparing outcomes after pancreatoduodenectomy. The observed differences can be used to guide collaborative initiatives to further improve the outcomes of pancreatic surgery.

Dataset Information

Comprehensive evaluation of composite gene features in cancer outcome prediction.

Publications

Comprehensive evaluation of composite gene features in cancer outcome prediction.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets