Present and former members of our group have developed several computational tools:
Profat, (Protein Function Annotation Tool), a novel software that predicts the function of unknown proteins by using a motif-profile based analysis, where repeat motifs sequences are used for protein family classification and later excluded based on a subsequent phylogenetic analysis on sequence similarities within protein families.
CandidateBacon is a tool that estimates the degree of separation between two genes in a network and validate that proposed interaction is not a statistical artifact.
Install the CandidateBacon package in R using the following commands:
download.file('http://www.computationalgenetics.se/CandidateBacon/CandidateBacon_1.0.tar.gz', destfile= "CandidateBacon_1.0.tar.gz") install.packages("CandidateBacon_1.0.tar.gz", type = "source")
Alternatively, the R package source can be downloaded [here]
A small example network is available in the “/data” folder in the CandidateBacon package installation directory.
A tool for assessing functional effects of amino acid substitutions based on physicochemical properties and evolutionary conservation. The tool is implemented in Python and is particularly suitable for analysis based on reference sequences from evolutionary distant species. The tool is available for Linux/Unix/OsX. [Download PASE]
MAPfastR is a fast and comprehensive software package for analyzing QTL data from outbred line-crosses that is developed for flexible analyses of large datasets. [Users' Forum]
Install the MAPfastR package in R using the following command:
Find MAPfastR tutorial [here].
DIPT (Detecting Interaction and Pathway tool) integrates the information from BioGRID and KEGG into a useful candidate gene screening tool. The online version is available here: http://computationalgenetics.se/DIPT/
A batch mode command line version is available for download below:
vGWAS – Variance Genome-wide Association
The package provides functions for genome-wide association using nonparametric variance test and some other further visualization and calculation. [Find it on R-Forge]
triM – tracing inheritance with Markov models
This is a method for calculating line origin probabilities for outbred line crosses that can be used for QTL analysis as described in Crooks et al. (in press) Genes, Genomes, Genetics. triM is particularly useful for large marker datasets where individual markers have low information, e.g. from SNP chips. It is implemented in in the codebase cnF2freq. [Download triM source code and OsX version] [Download triM Windows version] [cnF2freq]
qtl.outbred – Interface for Genotype Probabilities from Outbred Intercross
The package calculates and imports genotype probabilities from outbred intercross designs into an R object. This allows fast scans for main effect QTL and two way interactions. [find it on R-Forge]
MCIBD – Monte Carlo Identity-By-Descent Matrix Estimation
This is a package implemented for flexible identity-by-descent (IBD) matrix estimation in F2 pedigrees. The IBD matrix estimate is approached by a Monte Carlo strategy, where the segregation of founder alleles is easy to define. Epistatic IBD matrices can also be calculated for two arbitrary loci. Such IBD matrix estimates are typically used in variance component quantitative trait loci (QTL) analysis. Parallelization is implemented to enhance the performance of Monte Carlo sampling. [find it on R-Forge]
NOIA (Natural and Orthogonal Interactions) is a model of genetic effects. Such models transform genotypic values (the expected phenotypes for each of the genotypes) into parameters of insightful biological meaning. Examples are quantifications of how much allele substitutions (whether individual or population based) would affect phenotypes, what is (if any) the interaction effect between and among alleles or how well individuals would perform as breeders. [more on NOIA website]