Major topics of research.
1. Clustering: QC and DQC
We have developed the Quantum Clustering (QC) method in NIPS01 and PRL 2002.
This has since been applied to various problems, mostly in bioinformatics, several of
which are listed below. The algorithm is incorporated in the Matlab program
COMPACT that can be downloaded from the Research section of my website.
Recently we have developed the Dynamic Quantum Clustering method (PRL 2009). It
can be tested on the webtool http://dqc.cs.tau.ac.il/dqc which runs a MapleNet
application (only one user at a time…).
References (downloadable from the section Publications my website):
The Method of Quantum Clustering.
(David Horn and Assaf Gottlieb)
in proceedings of NIPS*01
Algorithm for data clustering in pattern recognition problems based on quantum
mechanics
(David Horn and Assaf Gottlieb)
Phys. Rev. Lett. 88 (2002) 18702
Novel clustering algorithm for microarray expression data in a truncated SVD space
(David Horn and Inon Axel)
Bioinformatics 19, 1110-1115, 2003.
COMPACT: A Comparative Package for Clustering Assessment
(Roy Varshavsky, Michal Linial and David Horn)
in G. Chen, Y. Pan, M. Guo and J. Lu (Eds.): Lecture Notes in Computer Science
3759 (2005) 159-167 Springer.
Novel Unsupervised Feature Filtering of Biological Data
(Roy Varshavsky, Assaf Gottlieb, Michal Linial and David Horn)
oral presentation at ISMB 2006, Bioinformatics 22(14):e507-513.
Global Considerations in Hierarchical Clustering Reveal Meaningful Patterns in Data.
(Roy Varshavshy, David Horn and Michal Linial) PLoSOne 2008
Dynamic quantum clustering: a method for visual exploration of structures in data.
(Marvin Weinstein and David Horn) Physical Review E 2009 (80) 066117
2. Bioinformatics: predicting function from sequence motifs of enzymes
It started from applying MEX, a motif extraction algorithm (PNAS 2005) to protein
sequences. We have applied it to all enzymes (Plos compbio 2007), deriving motifs
that are specific to EC categories, called Specific Peptides. We have shown their
connection to relevant biological markers (Proteins 2008). We have built an enzymes
prediction scheme called DME (BMC Bioinformatics 2009). A webtool is available at
http://adios.tau.ac.il/DME11.html. Recently we have applied SP searches directly to
short read metagenomic data.
Another application of MEX is creating common peptides from particular families of
proteins. We have looked for evolutionary patters using such analyses in Olfactory
Receptors and in aminoacyl tRNA synthetases.
References (downloadable from the section Publications on my website):
Unsupervised learning of natural languages
(Zach Solan, David Horn, Eytan Ruppin and Shimon Edelman)
Proc. Natl. Acad. Sc. 102 (2005) 11629-11634
Functional representation of enzymes by specific peptides
(Vered Kunik, Yasmine Meroz, Zach Solan, Ben Sandbank, Uri Weingart, Eytan
Ruppin and David Horn) PLOS Computational Biology 2007, 3(8):e167.
Biological roles of specific peptides in enzymes.
(Yasmine Meroz and David Horn) Proteins: Structure, Function, and Bioinformatics
72 (2), 606-612, 2008
Common peptides shed light on evolution of Olfactory Receptors
(Assaf Gottlieb, Tsviya Olender, Doron Lancet and David Horn)
BMC Evolutionary Biology 2009, 9:91
Data mining of enzymes using specific peptides.
(Uri Weingart, Yair Lavi and David Horn) BMC Bioinformatics 2009, 10:446
Deriving Enzymatic and Taxonomic Signatures of Metagenomes from Short
Read Data (Uri Weingart, Erez Persi, Uri Gophna and David Horn) BMC
Bioinformatics 2010, 11:390
Common Peptides Study of Aminoacyl-tRNA Synthetases.
(Assaf Gottlieb, Milana Frenkel-Morgenstern, Mark Safro and David Horn) PLoS
ONE 2011, 6(5): e20361. doi:10.1371/journal.pone.0020361
Peptide Markers of Aminoacyl tRNA Synthetases Facilitate Taxa Counting in
Metagenomic Data (Erez Persi, Uri Weingart, Shiri Freilich and David Horn) BMC
Genomics 2012, 13:65