Maximum Likelihood Inference on Multi-State Trees (ML.MSBD)
Inference of a multi-states birth-death model from a phylogeny, comprising a number of states N, birth and death rates for each state and on which edges each state appears. Inference is done using a hybrid approach: states are progressively added in a greedy approach. For a fixed number of states N the best model is selected via maximum likelihood. Reference: J. Barido-Sottani and T. Stadler (2017) <doi:10.1101/215491>.

Nearest Neighbor Based Multiple Imputation for Survival Data with Missing Covariates (NNMIS)
Imputation for both missing covariates and censored observations (optional) for survival data with missing covariates by the nearest neighbor based multiple imputation algorithm as described in Hsu et al. (2006) <doi:10.1002/sim.2452>, Long et al. (2012) <doi:10.5705/ss.2010.069>, Hsu et al. (2014) <doi:10.1080/10543406.2014.888444>, and Hsu and Yu (2017) <arXiv:1710.04721>. Note that the current version can only impute for a situation with one missing covariate.

Weight of Evidence Based Segmentation of a Variable (woeR)
Segment a numeric variable based on a dichotomous dependent variable by using the weight of evidence (WOE) approach (Ref: Siddiqi, N. (2006) <doi:10.1002/9781119201731.biblio>). The underlying algorithm adopts a recursive approach to create segments that are diverse in respect of their WOE values and meet the demands of user-defined parameters. The algorithm also aims to maintain a monotonic trend in WOE values of consecutive segments. As such, it can be particularly helpful in improving robustness of linear and logistic regression models.

A Shiny User Interface of Time Warping for Improved Gradient Matching (shinyKGode)
Interactive shiny application to perform inference of non-linear differential equations via gradient matching. Three (Lotka-Volterra, Fitz hugh Nagumo, and Biopathway) pre-defined models are provided, and users can also load their own models (in the Systems Biology Markup Language format) into the application.

Diagnostics for Pharmacometric Models (xpose)
Diagnostics for non-linear mixed-effects (population) models from ‘NONMEM’ <http://…/>. ‘xpose’ facilitates data import, creation of numerical run summary and provide ‘ggplot2’-based graphics for data exploration and model diagnostics.

High-Dimensional Mediation Analysis (HIMA)
Allows to estimate and test high-dimensional mediation effects based on sure independent screening and minimax concave penalty techniques. A joint significance test is used for mediation effect. Haixiang Zhang, Yinan Zheng, Zhou Zhang, Tao Gao, Brian Joyce, Grace Yoon, Wei Zhang, Joel Schwartz, Allan Just, Elena Colicino, Pantel Vokonas, Lihui Zhao, Jinchi Lv, Andrea Baccarelli, Lifang Hou, Lei Liu (2016) <doi:10.1093/bioinformatics/btw351>.

Advertisements