Dataset Information

Bayes optimal informer sets for early-stage drug discovery.

ABSTRACT: An important experimental design problem in early-stage drug discovery is how to prioritize available compounds for testing when very little is known about the target protein. Informer-based ranking (IBR) methods address the prioritization problem when the compounds have provided bioactivity data on other potentially relevant targets. An IBR method selects an informer set of compounds, and then prioritizes the remaining compounds on the basis of new bioactivity experiments performed with the informer set on the target. We formalize the problem as a two-stage decision problem and introduce the Bayes Optimal Informer SEt (BOISE) method for its solution. BOISE leverages a flexible model of the initial bioactivity data, a relevant loss function, and effective computational schemes to resolve the two-step design problem. We evaluate BOISE and compare it to other IBR strategies in two retrospective studies, one on protein-kinase inhibition and the other on anticancer drug sensitivity. In both empirical settings BOISE exhibits better predictive performance than available methods. It also behaves well with missing data, where methods that use matrix completion show worse predictive performance.

SUBMITTER: Yu P

PROVIDER: S-EPMC9376199 | biostudies-literature | 2023 Jun

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Bayes optimal informer sets for early-stage drug discovery.

Yu Peng P Ericksen Spencer S Gitter Anthony A Newton Michael A MA

Biometrics 20220325 2

An important experimental design problem in early-stage drug discovery is how to prioritize available compounds for testing when very little is known about the target protein. Informer-based ranking (IBR) methods address the prioritization problem when the compounds have provided bioactivity data on other potentially relevant targets. An IBR method selects an informer set of compounds, and then prioritizes the remaining compounds on the basis of new bioactivity experiments performed with the inf ...[more]

PMID: 35165892

Similar Datasets

Project description:BackgroundThere is no clear consensus regarding systemic treatment of early-stage ovarian cancer (OC). Clinical trials are challenging because of the relatively low incidence and good prognosis. Initial results of the International Collaborative Ovarian Neoplasm (ICON)1 trial demonstrated benefit in both overall survival (OS) and recurrence-free survival (RFS) with adjuvant chemotherapy. We report results of 10-year follow-up to establish whether benefits are maintained longer term and discuss how this and other available evidence from randomised trials can be used to guide treatment options regarding the need for, and choice of, adjuvant chemotherapy regimen.Patients and methodsICON1 recruited women with OC following primary surgery in whom there was uncertainty as to whether adjuvant chemotherapy was indicated. Patients were randomly assigned to adjuvant or no adjuvant chemotherapy. Platinum-based chemotherapy was recommended and 87% received single-agent carboplatin. Analyses of long-term treatment benefits and interaction with risk groups were carried out. A high-risk group of women was defined with stage 1B/1C grade 2/3, any stage 1 grade 3 or clear-cell histology.ResultsWith a median follow-up of 10 years, the estimated hazard ratio (HR) for RFS was 0.69 [95% confidence interval (CI) 0.51-0.94, P = 0.02] and OS 0.71 (95% CI 0.52-0.98, P = 0.04) in favour of chemotherapy. In absolute terms, there was a 10% (60%-70%) improvement in RFS and a 9% (64%-73%) improvement in OS; the benefit of chemotherapy might be greater in high-risk disease (18% improvement in OS). Uncertainty remains about the optimal chemotherapy regimen. The only randomised trial data available are from a subset of 120 stage 1 patients in ICON3 where the treatment difference, comparing carboplatin with carboplatin/paclitaxel was estimated with relatively wide CIs [progression-free survival HR = 0.71 (95% CI 0.39-1.32) and OS HR = 0.98 (95% CI 0.49-1.93)].ConclusionsExtended follow-up from ICON1 confirms that adjuvant chemotherapy should be offered to women with early-stage OC, particularly those with high-risk disease.Clinical trial numbersISRCTN11916376 for ICON1 and ISRCTN57157825 for ICON3.

Dataset Information

Bayes optimal informer sets for early-stage drug discovery.

Publications

Bayes optimal informer sets for early-stage drug discovery.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets