Everything here is legacy software from my research years — tools built for motif discovery, miRNA prediction and the logical modelling of gene regulatory networks. None of it is maintained any longer; what remains is kept as it was, for the record.
Parameter-free, heuristic method for the extraction of conserved patterns (motifs) in a set of DNA input sequences. Command-line program taking FASTA input; optional parameters (lambda, epsilon, sieve, mindist, maxdist) focus the search.
Legacy — unmaintained. Cite: Mendes et al., Bioinformatics 22(24): 2996–3002, 2006.
Framework to predict pre-miRNA candidates in single genomes, with optional conservation analysis in phylogenetically close species. Can incorporate RNA-seq data to predict mature sequences and estimate expression across conditions; candidates are evaluated by expression score, filtered with genome annotation when available, and analysed with stem-loop structural methods for metazoan genomes.
Legacy — unmaintained. The project website is currently inaccessible.
Two methods for identifying the set of attractors reachable from a given initial condition, and evaluating their reachability probability, over asynchronous discrete dynamics. Firefront is a quasi-exact method that follows all concurrent trajectories simultaneously while propagating state probabilities; Avatar is a modified Monte Carlo algorithm that handles strongly connected components and avoids revisiting states by rewiring the state transition graph.
Legacy — unmaintained. Published as Mendes et al., Frontiers in Physiology, 2018; implemented as new functionality of GINsim 3.0.
Composition and abstraction support for logical regulatory networks inside GinSIM, a tool for the analysis and specification of logical regulatory networks — the software behind the multicellular-systems work.
Legacy — unmaintained. Associated paper: Mendes et al., Bioinformatics, 2013.
DISCOVERER provides tools to identify complex motifs over-represented in the promoter regions of co-regulated genes. It is based on the MUSA and RISO algorithms: given a list of genes, it identifies over-represented motifs and compares them with transcription factor binding sites described in the YEASTRACT database.
Legacy — unmaintained. Associated paper: Monteiro, Mendes et al., Nucleic Acids Research, 2007.