Knowledge Discovery by Accuracy Maximization

A self-guided, weakly supervised learning algorithm for feature extraction from noisy and high-dimensional data. It facilitates the identification of patterns that reflect underlying group structures across all samples in a dataset. The method incorporates a novel strategy to integrate spatial information, improving the interpretability of results in spatially resolved data.

Chemical Structural Similarity Analysis

A new pipeline to explore chemical structural similarity across metabolites. It allows the metabolite classification in structurally-related modules and identifies common shared functional groups. The KODAMA algorithm is used to highlight structural similarity between metabolites. See Cacciatore S, Tenori L, Luchinat C, Bennett PR, MacIntyre DA. (2017) Bioinformatics , Cacciatore S, Luchinat C, Tenori L. (2014) Proc Natl Acad Sci USA , and Abdel-Shafy EA, Melak T, MacIntyre DA, Zadra G, Zerbini LF, Piazza S, Cacciatore S. (2023) Bioinformatics Advances .

Biodiversity Assessment Tools

Includes algorithms to assess alpha and beta diversity in all their dimensions (taxonomic, phylogenetic and functional). It allows performing a number of analyses based on species identities/abundances, phylogenetic/functional distances, trees, convex-hulls or kernel density n-dimensional hypervolumes depicting species relationships. Cardoso et al. (2015) .

Gene-Based Association Tests using the Actual Impurity Reduction (AIR) Variable Importance

Gene-based association tests using the actual impurity reduction (AIR) variable importance. The function aggregates AIR importance measures from a group of SNPs or probes and outputs a p-value for each gene. The procedures builds upon the method described in and will be published soon.

Research Assessment Tools

Includes algorithms to assess research productivity and patterns, such as the h-index and i-index. Cardoso et al. (2022) Cardoso, P., Fukushima, C.S. & Mammola, S. (2022) Quantifying the internationalization and representativeness in research. Trends in Ecology and Evolution, 37: 725-728.

A Fast Implementation of Partial Least Square

An implementation in 'Rcpp' / 'RcppArmadillo' of Partial Least Square algorithms. This package includes other functions to perform the double cross-validation and a fast correlation.

Linear Dynamical System Reconstruction

Streamflow (and climate) reconstruction using Linear Dynamical Systems. The advantage of this method is the additional state trajectory which can reveal more information about the catchment or climate system. For details of the method please refer to Nguyen and Galelli (2018) .

Develop a 'Telegram Bot' with R

Provides a pure interface for the 'Telegram Bot API' < http://core.telegram.org/bots/api>. In addition to the pure API implementation, it features a number of tools to make the development of 'Telegram' bots with R easy and straightforward, providing an easy-to-use interface that takes some work off the programmer.

Self Organising Maps for the Analysis of Molecular Dynamics Data

Processes data from Molecular Dynamics simulations using Self Organising Maps. Features include the ability to read different input formats. Trajectories can be analysed to identify groups of important frames. Output visualisation can be generated for maps and pathways. Methodological details can be found in Motta S et al (2022) . I/O functions for xtc format files were implemented using the 'xdrfile' library available under open source license. The relevant information can be found in inst/COPYRIGHT.

Create Interactive Web Maps Using 'The National Map' Services

Creates interactive web maps using the 'JavaScript' 'Leaflet' library with base layers of 'The National Map' ('TNM'). 'TNM' services provide access to base geospatial information that describes the landscape of the United States and its territories. This package is dependent on, and intended to be used with, the 'leaflet' package.