Projects and Software
Population Genetics
HAPAA:
a tool for ancestral haploblock reconstruction. Specifically, given
the genotype (for instance, as derived by an Illumina
genotyping array)
of an individual of admixed ancestry,
find the source population for each segment of the individual's
genome.
Protein Interaction Networks
A tool for aligning multiple global
protein interaction networks; Graemlin also supports search for
homology between a query
module of proteins and a database of
interaction networks.
Machine Learning
RNA Structure Prediction Protein Alignment Motif Finding Genomic Alignment Microarray Analysis
CONTRA: Conditionally trained models for
sequence analysis. See
CONTRAlign, a
protein sequence aligner with very high accuracy, especially in
twilight alignments. See
CONTRAfold, an
RNA secondary structure prediction tool. Stay
tuned for more...
CONTRAfold: Prediction of RNA secondary structure
with a Conditional Log-Linear model that relies on automatically
trained parameters, rather than on a physics-based energy model of
RNA folding.
CONTRAlign: A protein sequence aligner that users
can optionally train on feature sets such as secondary structure and
solvent accessibility; see the CONTRA project above.
A protein multiple sequence aligner
that exhibits high accuracy on popular benchmarks.
A protein multiple aligner that
automatically finds domain structures of sequences with shuffled and
repeated domain architectures.
MotifCut: a non-parametric
graph-based motif finding algorithm.
MotifScan: a non-parametric
method for representing motifs and scanning DNA sequences for known
motifs.
CompareProspector: motif finding
with Gibbs sampling & alignment.
Stanford ENCODE: Multiple Alignments
of 1% of the Human genome.
Typhon:
BLAST-like sequence search to a multiple alignments database.
LAGAN: tools for genomic alignment.
These include the MLAGAN multiple alignment tool, and Shuffle-LAGAN
for alignment with rearrangements.
Application of Independent Component Analysis (ICA) to microarrays.