Data processing: artificial intelligence – Plural processing systems
Reexamination Certificate
2000-04-14
2003-09-23
Davis, George B. (Department: 2121)
Data processing: artificial intelligence
Plural processing systems
C706S011000
Reexamination Certificate
active
06625585
ABSTRACT:
COPYRIGHT
A portion of the disclosure of this patent document contains material that is subject to copyright protection. The copyright owner has no objection to the facsimile reproduction by anyone of the patent disclosure, as it appears in the Patent and Trademark Office patent files or records, but otherwise reserves all copyright rights whatsoever.
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to computer-based analysis of data and to computer-based correlation of data features with data responses, in order to determine or predict which features correlate with or are likely to result in one or more responses. The invention is particularly suitable for use in the fields of chemistry, biology and genetics, such as to facilitate computer-based correlation of chemical structures with observed or predicted pharmacophoric activity. More particularly, the invention is useful in facilitating identification and development of potentially beneficial new drugs.
2. Description of Related Art
The global biotech and pharmaceutical industry is a $200 billion/year business. Most of the estimated $13 billion R&D spending in this industry is focused on discovering and developing prescription drugs. Current R&D effort is characterized by low drug discovery rates and long time-to-market.
In an effort to accelerate drug discovery, biotech and pharmaceutical firms are turning to robotics and automation. The old methods of rationally designing molecules using known structural relationships are being supplanted by a shotgun approach of rapidly screening it hundreds of thousands of molecules for biological activity. High Throughput Screening (HTS) is being used to test large numbers of molecules for biological activity. The primary goal is to identify hits or leads, which are molecules that affect a particular biological target in the desired manner. For instance and without limitation, a lead may be a chemical structure that binds particularly well to a protein.
Automated HTS systems are large, highly automated liquid handling and detection systems that allow thousands of molecules to be screened for biological activity against a test assay. Several pharmaceutical and biotech companies have developed systems that can perform hundreds of thousands of screens per day.
The increasing use of HTS is being driven by a number of other developments in the industry. The greater the number and diversity of molecules that are run through screens, the more successful HTS is likely to be. This fact has propelled rapid developments in molecule library collection and creation. Combinatorial chemistry systems have been developed that can automatically create hundreds of thousands of new molecules. Combinatorial chemistry is performed in large automated systems that are capable of synthesizing a wide variety of small organic molecules using combinations of “building block” reagents. HTS systems are the only way that the enormous volume of new molecules generated by combinatorial chemistry systems can be tested for biological activity. Another force driving the increased use of HTS is the Human Genome program and the companion field of bioinformatics that is enabling the rapid identification of gene function and accelerating the discovery of therapeutic targets. Companies do not have the resources to develop an exhaustive understanding of each potential therapeutic target. Rather, pharmaceutical and biotech companies use HTS to quickly find molecules that affect the target and may lead to the discovery of a new drug.
High throughput screening does not directly identify a drug. Rather the primary role of HTS is to detect lead molecules and supply directions for their optimization. This limitation exists because many properties critical to the development of a successful drug cannot be assessed by HTS. For example, HTS cannot evaluate the bioavailability, pharmacokinetics, toxicity, or specificity of an active molecule. Thus, further studies of the molecules identified by HTS are required in order to identify a potential lead to a new drug.
The further study, a process called lead discovery, is a time- and resource-intensive task. High throughput screening of a large library of molecules typically identifies thousands of molecules with biological activity that must be evaluated by a pharmaceutical chemist. Those molecules that are selected as candidates for use as a drug are studied to build an understanding of the mechanism by which they interact with the assay. Scientists try to determine which molecular properties correlate with high activity of the molecules in the screening assay. Using the drug leads and this mechanism information, chemists then try to identify, synthesize and test molecules analogous to the leads that have enhanced drug-like effect and/or reduced undesirable characteristics in a process called lead optimization. Ideally, the end result of the screening, lead discovery, and lead optimization is the development of a new drug for clinical testing.
As the number of molecules in the test library and the number of therapeutic target assays exponentially increase, lead discovery and lead optimization have become the new bottleneck in drug discovery using HTS systems. Because of the large number of HTS results that must be analyzed, scientists often seek only first-order results such as the identification of molecules in the library that exhibit high assay activity. In one such method, for instance, all of the molecules in the data set are divided into groups based on common properties of their molecular structures. An analysis is then made to determine which groups contain molecules with high activity levels and which groups contain molecules with low activity levels. Those groups representing high activity levels are then deemed to be useful groups. Commonly, the analysis will stop at this point, leaving chemists to analyze the members of the active groups in search of new or optimized leads.
In another method, a more extensive automated analysis is conducted in an effort to partition the molecules into groups of particular interest and particularly to derive structure-activity relationship rules. For instance, well known recursive partitioning techniques, commonly referred to as classification trees, may be used to iteratively partition a data set (such as results of HTS or other automated chemical synthesis) into active classes. The data set includes molecules and indicia of empirically determined potency (activity-level) per molecule.
According to this method, a set of descriptors is first generated, each indicating a structural feature that can be described as present or absent in a given molecule. For each molecule, a bit string is then built, indicating whether the molecule has each particular descriptor (1-bit) or not (0-bit). These strings are then configured as a matrix, in which each row represents a molecule and each column represents a descriptor. Recursive partitioning is then used to divide the molecules (rows) into exactly two groups according to whether the molecules have a particular “best” descriptor in common. The “best” descriptor is the descriptor that would result in the largest possible difference in average potency between those molecules containing the descriptor and those molecules not containing the descriptor.
The recursive partitioning method then continues iteratively with respect to each subdivided group, dividing each group into two groups based on a next “best” descriptor. The result of this process is a tree structure, in which some terminal nodes may contain a preponderance of inactive molecules (or molecules having relatively low potency) and other terminal nodes may contain a preponderance of active molecules (or molecules having relatively high potency) (the latter being “good terminal nodes”). Tracing the lineage of the structures defined by a good terminal node may then reveal molecular components that cooperatively reflect a high likelihood of potency.
Unfortunately, the use of recursive partitioning to partition molecul
MacCuish John D.
Nicolaou Christodoulos A.
Bioreason, Inc.
Davis George B.
McDonnell & Boehnen Hulbert & Berghoff
LandOfFree
Method and system for artificial intelligence directed lead... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and system for artificial intelligence directed lead..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and system for artificial intelligence directed lead... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3001597