Other

Eliza Corporation

Wyeth

Robert L Powers

John P Kroeker

Kevin Delos Parris

Stephen G Ferris

James Ming Chen

Mark L Stahl

William Stuart Somers

Laura Lin

Dominick Mobilio

Guang-Yi Xu

Amy Szepui Tam

Zhang Bao Xu

Terry Lindh

Terry K Lindh

Franklin J Moy

The present invention relates to the three dimensional structure of human collagenase 3 (MMP-13), as well as to (i) methods of using the MMP-13 structure to rationally design or identify compounds or molecules that inhibit or activate MMP-13 activity, and (ii) compounds identified using said methods.

Methods for designing agents that interact with MMP-13

This invention is directed to the crystal structure of Acyl Carrier Protein Synthase (ACPS) complexed with Acyl Carrier Protein (ACP), the solution structure of ACP, and to the use of these structures in rational drug design methods to identify agents that may interact with active sites of ACPS and ACP, and to the testing of these agents to identify agents that may represent novel antibiotics.

Methods for identifying agents that interact with an active site of acyl carrier protein synthase-acyl carrier protein complex

A fingerprint image is processed from its original gray tone image to a binary (black and white) image. The 'ridge' skeletal image is formed of the black portions of the binary image, being derived from the ridges of the fingerprint. A separate 'valley' skeletal image is formed of the white portions of the binary image, being derived from the valleys interposed between the ridges. Ridge minutiae are detected from the ridge skeletal image and processed to eliminate apparently inappropriate minutiae. Separately, valley minutiae are detected from the valley skeletal image and processed to eliminate apparently inappropriate minutia. Then the ridge and valley minutiae are processed together. Only those ridge minutiae that can be correlated locally with individual valley minutiae survive to be passed to a fingerprint matcher.

Ridge-valley minutia associator for fingerprints

A fingerprint matcher links together minutiae from a search print and minutiae from a file print in a chain to determine if the two prints match. Minutiae are extracted from the images of the search and file prints. Geometric and other relationships between close minutiae in each print are calculated. A table of candidate 'steps' is formed. Each step consists of four minutiae, two close search minutiae and two close file minutiae. Each 'rung' of the step is a search-file pair, the candidate mating of presumed identical minutiae in the two images. Each 'riser' of the step is formed by two close minutiae, search minutiae on the one side and file minutiae on the other. A candidate step is entered into the table if and only if the relationships between search minutiae are similar to those between file minutiae. The number of candidate steps is reduced by requiring that each step adhere to various rules regarding the presence of other steps in the table and the identity of minutiae therein. Maximally long chains of steps are formed which are called 'hyperladders', each of which links together surviving steps. The hyperladders are said to be 'hyper' because at each search-file pair rung a multiplicity of 'riser' pairs may emerge, each to another and different search-file pair 'rung'. Within the largest hyperladder formed from a given search-file pair of prints the number of links is a measure of the likelihood that the search and file fingerprints are the same, and an indication of a match is returned if the number of links exceeds a predetermined threshold value.

Hyperladder fingerprint matcher

A phoneme estimator in a speech-recognition system includes energy detect circuitry for detecting the segments of a speech signal that should be analyzed for phoneme content. Speech-element processors then process the speech signal segments, calculating nonlinear representations of the segments. The nonlinear representation data is applied to speech-element modeling circuitry which reduces the data through speech element specific modeling. The reduced data are then subjected to further nonlinear processing. The results of the further nonlinear processing are again applied to speech-element modeling circuitry, producing phoneme isotype estimates. The phoneme isotype estimates are rearranged and consolidated, that is, the estimates are uniformly labeled and duplicate estimates are consolidated, forming estimates of words or phrases containing minimal numbers of phonemes. The estimates may then be compared with stored words or phrases to determine what was spoken.

Speech recognition circuitry employing nonlinear processing speech

A phoneme estimator in a speech-recognition system includes energy detect circuitry for detecting the segments of a speech signal that should be analyzed for phoneme content. Speech-element processors then process the speech signal segments, calculating nonlinear (powers and products) representations of the segments. The nonlinear representation data is applied to speech-element modeling circuitry which reduces the data through speech element specific modeling. The reduced data are then subjected to further nonlinear processing. The results of the further nonlinear processing are again applied to speech-element modeling circuitry, producing phoneme isotype estimates. The phoneme isotype estimates are rearranged and consolidated, that is, the estimates are uniformly labeled and duplicate estimates are consolidated, forming estimates of words or phrases containing minimal numbers of phonemes. The estimates may then be compared with stored words or phrases to determine what was spoken.

Speech-recognition circuitry employing nonlinear processing, speech

A phoneme estimator (12) in a speech-recognition system (10) includes trigger circuitry (18, 22) for identifying the segments of speech that should be analyzed for phoneme content. Speech-element processors (24, 26, and 28) calculate the likelihoods that currently received speech contains individual phonemes, but they operate only when the trigger circuitry identifies such segments. The computation-intensive processing for determining phoneme likelihoods is thus performed on only a small subset of the received speech segments. The accuracy of the speech-element processors (24, 26, and 28) is enhanced because these processors operate by recognition of patterns not only in elements of the data-reduced representations of the received speech but also in higher-ordered products of those elements; that is, these circuits employ non-linear modeling for phoneme identification.

Growing community of inventors

Topsfield, MA, United States of America

Robert L Powers