Paul D. Scott

Dimerization of G-protein-coupled receptors.

The evolutionary trace (ET) method, a data mining approach for determining significant levels of amino acid conservation, has been applied to over 700 aligned G-protein-coupled receptor (GPCR) sequences. The method predicted the occurrence of functionally important clusters of residues on the external faces of helices 5 and 6 for each family or subfamily of receptors; similar clusters were observed on helices 2 and 3. The probability that these clusters are not random was determined using Monte Carlo techniques. The cluster on helices 5 and 6 is consistent with both 5,6-contact and 5,6-domain swapped dimer formation; the possible equivalence of these two types of dimer is discussed because this relates to activation by homo- and heterodimers. The observation of a functionally important cluster of residues on helices 2 and 3 is novel, and some possible interpretations are given, including heterodimerization and oligomerization. The application of the evolutionary trace method to 113 aligned G-protein sequences resulted in the identification of two functional sites. One large, well-defined site is clearly identified with adenyl cyclase, beta/gamma and regulator of G-protein signaling (RGS) binding. The other G-protein functional site, which extends from the ras-like domain onto the helical domain, has the correct size and electrostatic properties for GPCR dimer binding. The implications of these results are discussed in terms of the conformational changes required in the G-protein for activation by a receptor dimer. Further, the implications of GPCR dimerization for medicinal chemistry are discussed in the context of these ET results.

The role of forgetting in learning

This paper is a discussion of the relationship between learning and forgetting. An analysis of the economics of learning is carried out and it is argued that knowledge can sometimes have a negative value. A series of experiments involving a program which learns to traverse state spaces is described. It is shown that most of the knowledge acquired is of negative value even though it is correct and was acquired solving similar problems. It is shown that the value of the knowledge depends on what else is known and that random forgetting can sometimes lead to substantial improvements in performance. It is concluded that research into knowledge acquisition should take seriously the possibility that knowledge may sometimes be harmful. The view is taken that learning and forgetting are complementary processes which construct and maintain useful representations of experience. Research on machine learning is concerned with the problem of how a system may acquire knowledge that it does not possess. It is therefore not surprising that relatively little attention has been paid to the converse problem: How may a system dispose of knowledge it already possess? This is the phenomenon that is termed forgetting when it occurs in humans, and is usually regarded as an unfortunate failure of the memory system. It is our contention that this negative view of forgetting is misplaced and that far from being a shortcoming it is a very useful process which facilitates effective knowledge acquisition. Learning is a process in which an organized representation of experience is constructed (Scott 1983). Forgetting is a process in which parts of that organized representation are rearranged or dismantled. The two processes are thus complementary and the resulting representation is the joint product of both. Mechanisms of forgetting therefore merit study alongside those of acquisition since it is the two together which constitute learning. Our notion of forgetting is fairly broad. In addition to the obvious mechanism of deletion of items of knowledge it also includes changes in the knowledge structure which render particular items relatively or completely inaccessible. It thus includes processes which weaken memory traces or isolate fragments of a knowledge base. Such changes can be viewed as partial removals with deletion as a limiting case which produces complete removal. In this paper we attempt to explore the role of forgetting in machine learning systems. We begin by discussing the circumstances in which it is better to dispose of an item of knowledge than retain it. Then we describe some experimental work we have done in order to demonstrate that even correct knowledge acquired in the course of solving similar problems can be a disadvantage to a system. 2. The Economics of Learning

Information Filtering: Selection Mechanisms in Learning Systems

Knowledge has traditionally been considered to have a beneficial effect on the performance of problem solvers but recent studies indicate that knowledge acquisition is not necessarily a monotonically beneficial process, because additional knowledge sometimes leads to a deterioration in system performance. This paper is concerned with the problem of harmful knowledge: that is, knowledge whose removal would improve a systems performance. In the first part of the paper a unifying framework, called the information filtering model, is developed to define the various alternative methods for eliminating such knowledge from a learning system where selection processes, called filters, may be inserted to remove potentially harmful knowledge. These filters are termed selective experience, selective attention, selective acquisition, selective retention, and selective utilization. The framework can be used by developers of learning systems as a guide for selecting an appropriate filter to reduce or eliminate harmful knowledge.In the second part of the paper, the framework is used to identify a suitable filter for solving a problem caused by the acquisition of harmful knowledge in a learning system called LASSY. LASSY is a system that improves the performance of a PROLOG interpreter by utilizing acquired domain specific knowledge in the form of lemmas stating previously proved results. It is shown that the particular kind of problems that arise with this system are best solved using a novel utilization filter that blocks the use of lemmas in attempts to prove subgoals that have a high probability of failing.

Coupling and cohesion measures for evaluation of component reusability

This paper provides an account of new measures of coupling and cohesion developed to assess the reusability of Java components retrieved from the internet by a search engine. These measures differ from the majority of established metrics in two respects: they reflect the degree to which entities are coupled or resemble each other, and they take account of indirect couplings or similarities. An empirical comparison of the new measures with eight established metrics shows the new measures are consistently superior at ranking components according to their reusability.

Ranking reusability of software components using coupling metrics

This paper provides an account of new static measures of coupling developed to assess the reusability of Java components retrieved from the internet by a search engine. These measures differ from the majority of established metrics in three respects: they take account of indirect coupling, they reflect the degree to which two classes are coupled, and they take account of the functional complexity of classes. An empirical comparison of the new measures with six established coupling metrics is described. The new measures are shown to be consistently superior at ranking components according to their reusability.

New Coupling and Cohesion Metrics for Evaluation of Software Component Reusability

An account of new measure of coupling and cohesion developed to assess the reusability of Java components is proposed in this paper. These measures differ from the majority of established metrics in two respects: they reflect the degree to which entities are coupled or resemble each other, and they take account of indirect couplings or similarities. An empirical comparison of the new measures with eight established metrics is described. The new measures are shown to be consistently superior at measuring component reusability.

Evaluating data mining procedures: techniques for generating artificial data sets

Abstract In this article, we discuss the need to evaluate the performance of data mining procedures and argue that tests done with real data sets cannot provide all the information needed for a thorough assessment of their performance characteristics. We argue that artificial data sets are therefore essential. After a discussion of the desirable characteristics of such artificial data, we describe two pseudo-random generators. The first is based on the multi-variate normal distribution and gives the investigator full control of the degree of correlation between the variables in the artificial data sets. The second is inspired by fractal techniques for synthesizing artificial landscapes and can produce data whose classification complexity can be controlled by a single parameter. We conclude with a discussion of the additional work necessary to achieve the ultimate goal of a method of matching data sets to the most appropriate data mining technique.

Information filters and their implementation in the SYLLOG system

Publisher Summary This chapter presents a general framework for the reduction of the harmfulness of learned knowledge. The study of knowledge has always been one of the central issues of the AI research. The potential harmfulness of correct knowledge has become a prominent concern alongside the harm because of incorrect knowledge. Knowledge is harmful if the costs associated with retaining it are greater than its benefits. Irrelevant knowledge and redundant knowledge are two types of knowledge that are very often harmful. When the knowledge is acquired by a learning program, it is desirable that the harmfulness of the knowledge will be eliminated or at least reduced by the program itself. Information in a learning system flows from the experiences that the system is facing, through the acquisition procedure to the knowledge base, and thence to the problem solver. An information filter is any process that removes information at any stage of this flow. The filters that are inserted between the experience space and the acquisition procedure data filters, and the filters that are inserted between the acquisition procedure and the problem solver knowledge filters. Information filters essentially are functions that are inserted between the input to the learning system and the input to the problem solver. The role of these functions is to eliminate (or reduce) harmful knowledge.

Entropy and oligomerization in GPCRs

Evolutionary trace (ET) and entropy are two related methods for analyzing a multiple sequence alignment to determine functionally important residues in proteins. In this article, these methods have been enhanced with a view to reinvestigate the issue of GPCR dimerization and oligomerization. In particular, cluster analysis has replaced the subjective visual analysis element of the original ET method. Previous applications of the ET method predicted two dimerization interfaces on the external transmembrane lipid-facing region of GPCRs; these were discussed in terms of dimerization and linear oligomers. Removing the subjective element of the ET method gives rise to the prediction of functionally important residues on the external face of each transmembrane helix for a large number of class A GPCRs. These results are consistent with a growing body of experimental information that, taken over many receptor subtypes, has implicated each transmembrane helix in dimeric interactions. In this application, entropy gave superior results to those obtained from the ET method in that its use gives rise to higher z-scores and fewer instances of z-scores below 3.

Experience Selection and Problem Choice in an Exploratory Learning System

A fully autonomous exploratory learning system must perform two tasks that are not required of supervised learning systems: experience selection and problem choice. Experience selection is the process of choosing informative training examples from the space of all possible examples. Problem choice is the process of identifying defects in the domain theory and determining which should be remedied next. These processes are closely related because the degree to which a specific experience is informative depends on the particular defects in the domain theory that the system is attempting to remedy. In this article we propose a general control structure for exploratory learning in which problem choice by an information-theoretic “curiosity” heuristic: the problem chosen then guides the selection of training examples. An implementation of an exploratory learning system based on this control structure is described, and a series of experimental results are presented.