ABSTRACT
MOTIVATION: The use of gene microchips has enabled a rapid accumulation of gene-expression data. One of the major challenges of analyzing this data is the diversity, in both size and signal strength, of the various modules in the gene regulatory networks of organisms. RESULTS: Based on the iterative signature algorithm [Bergmann,S., Ihmels,J. and Barkai,N. (2002) Phys. Rev. E 67, 031902], we present an algorithm-the progressive iterative signature algorithm (PISA)-that, by sequentially eliminating modules, allows unsupervised identification of both large and small regulatory modules. We applied PISA to a large set of yeast gene-expression data, and, using the Gene Ontology database as a reference, found that the algorithm is much better able to identify regulatory modules than methods based on high-throughput transcription-factor binding experiments or on comparative genomics.
Subject(s)
Algorithms , Gene Expression Profiling/methods , Gene Expression Regulation/physiology , Models, Genetic , Oligonucleotide Array Sequence Analysis/methods , Signal Transduction/physiology , Transcription Factors/metabolism , Computer Simulation , Models, Statistical , Pattern Recognition, Automated/methodsABSTRACT
Besides the usual conductance plateaus at multiples of 2e(2)/h, quantum point contacts typically show an extra plateau at approximately 0.7(2e(2)/h), believed to arise from electron-electron interactions that prohibit the two spin channels from being simultaneously occupied. We present evidence that the disappearance of the 0.7 structure at very low temperature signals the formation of a Kondo-like correlated spin state. Evidence includes a zero-bias conductance peak that splits in a parallel field, scaling of conductance to a modified Kondo form, and consistency between peak width and the Kondo temperature.
ABSTRACT
Protein structures are a very special class among all possible structures. It has been suggested that a "designability principle" plays a crucial role in nature's selection of protein sequences and structures. Here, we provide a theoretical base for such a selection principle, using a simple model of protein folding based on hydrophobic interactions. A structure is reduced to a string of 0s and 1s, which represent the surface and core sites, respectively, as the backbone is traced. Each structure is therefore associated with one point in a high dimensional space. Sequences are represented by strings of their hydrophobicities and thus can be mapped into the same space. A sequence that lies closer to a particular structure in this space than to any other structures will have that structure as its ground state. Atypical structures, namely those far away from other structures in the high dimensional space, have more sequences that fold into them and are thermodynamically more stable. We argue that the most common folds of proteins are the most atypical in the space of possible structures.
Subject(s)
Protein Folding , Proteins/chemistry , Models, Biological , Solvents , Structure-Activity Relationship , Surface PropertiesABSTRACT
The basic building blocks of an interconnection network are the switching elements. We examine two optical implementations for a basic switching element: (1) ferroelectric liquid-crystal light valves and (2) Fabry-Perot étalons. For these two examples we report a trade-off between the size, i.e., the number of input and output channels, and the switching speed. We speculate that it may be a general property of optical switching elements that size and speed cannot be optimized simultaneously.
