spatsurv: An R Package for Bayesian Inference with Spatial Survival Models

Mutational signature analysis aims to infer the mutational spectra and relative exposures of processes that contribute mutations to genomes. Different models for signature analysis have been developed, mostly based on non-negative matrix factorisation or non-linear optimisation. Here we present sigfit, an R package for mutational signature analysis that applies Bayesian inference to perform fitting and extraction of signatures from mutation data. We compare the performance of sigfit to prominent existing software, and find that it compares favourably. Moreover, sigfit introduces novel probabilistic models that enable more robust, powerful and versatile fitting and extraction of mutational signatures and broader biological patterns. The package also provides user-friendly visualisation routines and is easily integrable with other bioinformatic packages.

Download Full-text

Spatial survival modelling of business re-opening after Katrina: Survival modelling compared to spatial probit modelling of re-opening within 3, 6 or 12 months

Statistical Modelling ◽

10.1177/1471082x20967158 ◽

2020 ◽

pp. 1471082X2096715

Author(s):

Roger S. Bivand ◽

Virgilio Gómez-Rubio

Keyword(s):

Hurricane Katrina ◽

Survival Data ◽

R Package ◽

American Statistical Association ◽

Survival Models ◽

Royal Statistical Society ◽

Business Decisions ◽

Modeling And Analysis ◽

Probit Models ◽

Spatial Probit

Zhou and Hanson; Zhou and Hanson; Zhou and Hanson ( 2015 , Nonparametric Bayesian Inference in Biostatistics, pages 215–46. Cham: Springer; 2018, Journal of the American Statistical Association, 113, 571–81; 2020, spBayesSurv: Bayesian Modeling and Analysis of Spatially Correlated Survival Data. R package version 1.1.4) and Zhou et al. (2020, Journal of Statistical Software, Articles, 92, 1–33) present methods for estimating spatial survival models using areal data. This article applies their methods to a dataset recording New Orleans business decisions to re-open after Hurricane Katrina; the data were included in LeSage et al. (2011b , Journal of the Royal Statistical Society: Series A (Statistics in Society), 174, 1007—27). In two articles ( LeSage etal., 2011a , Significance, 8, 160—63; 2011b, Journal of the Royal Statistical Society: Series A (Statistics in Society), 174, 1007—27), spatial probit models are used to model spatial dependence in this dataset, with decisions to re-open aggregated to the first 90, 180 and 360 days. We re-cast the problem as one of examining the time-to-event records in the data, right-censored as observations ceased before 175 businesses had re-opened; we omit businesses already re-opened when observations began on Day 41. We are interested in checking whether the conclusions about the covariates using aspatial and spatial probit models are modified when applying survival and spatial survival models estimated using MCMC and INLA. In general, we find that the same covariates are associated with re-opening decisions in both modelling approaches. We do however find that data collected from three streets differ substantially, and that the streets are probably better handled separately or that the street effect should be included explicitly.

Download Full-text

BLNN: An R package for training neural networks using Bayesian inference

SoftwareX ◽

10.1016/j.softx.2020.100432 ◽

2020 ◽

Vol 11 ◽

pp. 100432

Author(s):

Taysseer Sharaf ◽

Theren Williams ◽

Abdallah Chehade ◽

Keshav Pokhrel

Keyword(s):

Neural Networks ◽

Bayesian Inference ◽

R Package

Download Full-text

Cost-effectiveness Analysis in R Using a Multi-state Modeling Survival Analysis Framework: A Tutorial

Medical Decision Making ◽

10.1177/0272989x16651869 ◽

2016 ◽

Vol 37 (4) ◽

pp. 340-352 ◽

Cited By ~ 19

Author(s):

Claire Williams ◽

James D. Lewsey ◽

Andrew H. Briggs ◽

Daniel F. Mackay

Keyword(s):

Cost Effectiveness ◽

Markov Property ◽

R Package ◽

Cost Effectiveness Analysis ◽

Sensitivity Analyses ◽

Survival Models ◽

Decision Analytic Model ◽

Modeling Approach ◽

Effectiveness Analysis ◽

Markov Decision

This tutorial provides a step-by-step guide to performing cost-effectiveness analysis using a multi-state modeling approach. Alongside the tutorial, we provide easy-to-use functions in the statistics package R. We argue that this multi-state modeling approach using a package such as R has advantages over approaches where models are built in a spreadsheet package. In particular, using a syntax-based approach means there is a written record of what was done and the calculations are transparent. Reproducing the analysis is straightforward as the syntax just needs to be run again. The approach can be thought of as an alternative way to build a Markov decision-analytic model, which also has the option to use a state-arrival extended approach. In the state-arrival extended multi-state model, a covariate that represents patients’ history is included, allowing the Markov property to be tested. We illustrate the building of multi-state survival models, making predictions from the models and assessing fits. We then proceed to perform a cost-effectiveness analysis, including deterministic and probabilistic sensitivity analyses. Finally, we show how to create 2 common methods of visualizing the results—namely, cost-effectiveness planes and cost-effectiveness acceptability curves. The analysis is implemented entirely within R. It is based on adaptions to functions in the existing R package mstate to accommodate parametric multi-state modeling that facilitates extrapolation of survival curves.

Download Full-text