Characterization of bio-chemical signals by inductive logic programming

doi:10.1016/S0950-7051(01)00129-0

Knowledge-Based Systems

Volume 15, Issues 1–2, January 2002, Pages 129-137

https://doi.org/10.1016/S0950-7051(01)00129-0 Get rights and content

Abstract

This paper presents a methodology to design a discrete-event system (DES) for the on-line supervision of a biotechnological process. The DES is synthesized applying wavelet transform and inductive logic programming on the measured signals constrained to the biotechnologist expert validation.

Introduction

Despite the complexity of biotechnological processes, biotechnologists are able to identify normal and abnormal situations, undertaking suitable actions accordingly. Process experts are able to draw such conclusions by analyzing a set of measured signals collected from the plant. The inexistence of satisfactory mathematical models impedes model-based approaches to be used in supervisory tasks, thus involving other strategies to be applied. That suggests that, in spite of the lack of process models, measured data can be used instead in the supervisory system development.

One of the main contribution of artificial intelligence (AI) to biological or chemical processes turns out to be the classification of an increasing amount of data. Can we do more than that and can an AI program contribute to help in the discovery of hidden rules in some process as complex as this. In fact, even if we can predict, for instance, mutagenicity of a given molecule or the secondary structure of proteins, with high degree of accuracy, this is not sufficient to give a deep insight of the observed behavior.

The main process we are concerned with in our paper is a bio-reaction, namely the dynamical behavior of yeast during chemostat cultivation. Starting from the observation of a set of evolutive parameters, our final aim is to extract logical rules to infer the physiological state of the yeast. Doing so, we obtain not only a better understanding of the system evolution but also the possibility to integrate the inferred rules in a full on-line control process. The first thing we have to do is to capture and analyze the parameters given by the sensors. These signals must be treated to be finally given to the logic machine. Thus, two things have to be done: firstly, to denoise the signals, secondly to compute the local maximum values of the given curves. In fact, we are more interested in the variations of the signals than in their pure instantaneous values. We use a method issued from wavelets theory [1] which tends to replace classical Fourier analysis. Among the advantages of WT, we can cite:

1.
The possibility to decompose a signal into multiple resolutions or scales. This property allows, for a given signal, to exhibit self-similarity at different magnifications.
2.
The localization of wavelets in both time and frequency allows to analyze local features (this is not the case for Fourier transformation).
3.
Wavelets have an excellent data compression rate.

At the end of this purely analytic treatment, we dispose of a set of clean values for each critical parameter.

Now, our idea is to apply inductive logic programming to exhibit, starting from a finite sample set of numerical observations, a number of logical formulae which organize the knowledge using causal relationships. Inductive logic programming is a sub-field of machine learning based upon a first-order logic framework. So, instead of giving a mathematical formula (for instance a differential equation) or a statistical prediction involving the different parameters, we provide a set of implicative logical formulae.

A part of these formulae can generally be inferred by a human expert, so it is a way to validate partially the mechanism. But it remains some new formulae which express an unknown causality relation: in that sense, this is a kind of knowledge discovery. As far as we know, one of the novelties of our work is the introduction of a time dimension to simulate the dynamic process. In logic, this time variable is in general not considered except with some specific modal logics. So, we modelize the time with an integer-valued variable (Fig. 1).

Section snippets

Yeast production application: example

Yeasts are a very well-studied micro-organisms [2] and today, such micro-organism as Saccharomyces cerevisiae which make the object of this study, are largely used in various sectors of the biomedical and biotechnology industrial processes.

Saccharomyces Cerevisiae is studied under oxidative regime (i.e. no ethanol production) to produce yeast under a laboratory environment in a bioreactor. Two different procedures are applied: a batch procedure that is followed by a continuous procedure. The

Statistical estimation

It is not new that noise generally affects the quality of the observed date. The standard model to described a noisy signal is: $X[n]=f[n]+W[n]$ where f is the determinist signal, and W[n] is a gaussian white noise of variance σ².

To estimate f by $f ̂ = f ̂ (X)$ two methods have been tested:

1.
linear methods: having the main defect that blur the shape of the signals,
2.
non-linear method: wavelet transform.

In signal classification it is important to find a representation of the signal which are translation

Knowledge based methodology

In learning there is a constant interaction between the creation and the recognition of concepts. The goal of the methodology is to obtain a model of the process, which can be used in a supervisory system for condition monitoring. The complexity of this model imposes the co-operation of data mining techniques along with the expect knowledge. When only expert knowledge is used to identify process situations or states, any of these situations can arise:

1.
the expert can express only a partial

Discussion and conclusion

In this paper, we apply logical tools to get explanation rules concerning the behavior of a bio-reactor. The ability to incorporate background knowledge and re-use past experiences marks out ILP as a very effective solution for our problem. Instead of simply giving classification results, we get some logical rules establishing a causality relationship between different parameters of the bio-machinery. Among these rules, some ones are validated by expert knowledge, but some new ones have been

References (6)

A. Arneodo, F. Argoul, E. Bacry, J. Elezgaray, J.F. Muzy. Ondelettes, Multifractales et Turbulence. Diderot Editeur,...
K. Konstantinov et al.
Physiological state control of fermentation processes
Biotechnology and Bioengineering
(1989)
J. Rocca. Technical report of Laboratoire d'automatisme et architecutre des sytemes,...

There are more references available in the full text version of this article.

Cited by (14)

DNA Double-Strand Break-Based Nonmonotonic Logic
2015, Emerging Trends in Computational Biology, Bioinformatics, and Systems Biology: Algorithms and Software Tools
Biological systems are complex systems with a dynamic evolution in terms of structure and biological causality. In order to model and control these systems, we must observe and simulate their behavior using relevant models. This chapter introduces a logical model based on default logic for DNA double-strand break (DSB) modeling. This new approach handles the uncertainties of knowledge and missing information concerning the conditions of activation/inhibition of some proteins. This model can be applied to any signaling pathway, even those with more than 6000 genes and proteins.
Bioprocess diagnosis based on the empirical use of distance measures in the theory of belief functions
2014, Engineering Applications of Artificial Intelligence
Citation Excerpt :
The event-tracking is classified as “normal” or “faulty” using a supervised classifier. Heuristics based expert systems rely on capturing knowledge and know-how) on the growth conditions of microorganism, and take into account the different fault occurrences and process variable interactions that the plant personnel can envision, by capturing them a rule base (Steyer, 1991; Doncescu et al., 2002). Model-based statistical signal processing: In these approaches, statistical models are developed using event-tracking data collected during the routine normal operation of fermentation.
Microorganisms plays a central role in the production of a wide range of industrial chemicals, enzymes and antibiotics. The rate of product formation in a given industrial process is directly related to the rate of biomass formation which is influenced directly or indirectly by a whole host of different environmental factors. In this paper we propose to use distance measures between basic belief assignment in the context of the belief functions theory, in order to diagnosis the relevance of bioprocess sensors and actors which measure the environmental factors.
A new relational Tri-training system with adaptive data editing for inductive logic programming
2012, Knowledge-Based Systems
Citation Excerpt :
By using first-order logic as the representational mechanism for hypotheses and examples, ILP (Inductive Logic Programming) can overcome the two main limitations of classical machine learning techniques (CN2 [1], HDT (Hybrid Decision Tree) [2]): (1) the use of a limited knowledge representation formalism (essentially a propositional logic), and (2) difficulties in using substantial background knowledge in the learning process [3]. In recent years, ILP has been applied successfully to a number of real-world fields, such as structure–activity for drug design [4], characterization of bio-chemical signals [5], gene regulation prediction [6] and protein–protein interactions [7,8]. ILP systems automate the construction of first-order definite clause theories from examples and background knowledge, its main task is to find a hypothesis H which covers all positive examples and none of the negative examples.
Relational Tri-training (R-Tri-training for short), as a relational semi-supervised learning system, can effectively exploit unlabeled examples to improve the generalization ability. However, the R-Tri-training may also suffer from the common problem in traditional semi-supervised learning, i.e., the performance is usually not stable for the unlabeled examples often be wrongly labeled and accumulated during the iterative learning process. In this paper, a new Relational Tri-training system named ADE-R-Tri-training (R-Tri-training with Adaptive Data Editing) is proposed. Not only does it employ a specific data editing technique to identify and correct the examples possibly mislabeled throughout the co-labeling iterations, but it also takes an adaptive strategy to decide whether to trigger the editing operation according to different cases. The adaptive strategy consists of five pre-conditional theorems, all of which ensure the iterative reduction of classification error under PAC (Probably Approximately Correct) learning theory. Experiments on well-known benchmarks show that ADE-R-Tri-training can more effectively enhance the performance of the hypothesis learned than R-Tri-training.
Improving the scalability of ILP-based multi-relational concept discovery system through parallelization
2012, Knowledge-Based Systems
Citation Excerpt :
Using logic in data mining is a common technique. It has applications in concept learning systems [4–7], association rule mining [8,9], handling aggregation and numerical data [10–13], coping with noisy and incomplete data [13–15], engineering applications [16,17], improving similarity measures in complex data [18]. Quinlan [4] defines the problem of concept learning as learning Horn clauses, disjunction of literals with exactly one positive literal being the concept relation, from relational data.
Due to the increase in the amount of relational data that is being collected and the limitations of propositional problem definition in relational domains, multi-relational data mining has arisen to be able to extract patterns from relational data. In order to cope with intractably large search space and still to be able to generate high-quality patterns, ILP-based multi-relational data mining and concept discovery systems employ several search strategies and pattern limitations. Another direction to cope with the large search space is using parallelization. By parallel data mining, improvement in time efficiency and scalability can be provided without further limiting the language patterns. In this work, we describe a method for concept discovery with parallelization on an ILP-based concept discovery system. The non-parallel algorithm, namely Concept Rule Induction System (CRIS), is modified in such a way that the parts that involve high amount of query processing, which causes bottleneck, are reorganized in a data parallel way. The resulting algorithm is called, Parallel CRIS (pCRIS). A set of experiments is conducted in order to evaluate the performance of the proposed method.
Concept discovery on relational databases: New techniques for search space pruning and rule quality improvement
2010, Knowledge-Based Systems
Multi-relational data mining has become popular due to the limitations of propositional problem definition in structured domains and the tendency of storing data in relational databases. Several relational knowledge discovery systems have been developed employing various search strategies, heuristics, language pattern limitations and hypothesis evaluation criteria, in order to cope with intractably large search space and to be able to generate high-quality patterns. In this work, we introduce an ILP-based concept discovery framework named Concept Rule Induction System (CRIS) which includes new approaches for search space pruning and new features, such as defining aggregate predicates and handling numeric attributes, for rule quality improvement. In CRIS, all target instances are considered together, which leads to construction of more descriptive rules for the concept. This property also makes it possible to use aggregate predicates more accurately in concept rule construction. Moreover, it facilitates construction of transitive rules. A set of experiments is conducted in order to evaluate the performance of proposed method in terms of accuracy and coverage.
Detection and characterization of physiological states in bioprocesses based on Hölder exponent
2008, Knowledge-Based Systems
Citation Excerpt :
However, no reliable technique exist to carry out real-time measurement of non-volatile substrates and metabolites in the fermentor. Several works using various approaches, lead to the conclusion that the limits of a state are linked to the singularities of biochemical signals: Steyer et al. [22] (using expert system and fuzzy logic), Bakshi and Stephanopoulos [1] (using expert system and wavelets) and Doncescu et al. [3] (using inductive logic) show that the beginning and the end of a state correspond to singularities of the biochemical signals measured during the process. In a fed-batch bioprocess, a physiological state can occur several times during the experience.
Today, the pace of progress in fermentation is fast and furious, particularly since the advent of genetic engineering and the recent advances in computer sciences and process control. The high cost associated with many fermentation processes makes optimization of bioreactor performance trough command control very desirable. Clearly, control of fermentation is recognized as a vital component in the operation and successful production of many industries. Today’s advances in measurement, data acquisition and handling technologies provide a wealth of new data which can be used to improve existing models. In this article we propose a method of physiological state identification based on segmentation of bioreactor sensors signals. The underlying of this method is based on the detection of signals singularities by the Maximum of Modulus of Wavelets Transform and their characterization by Hölder exponent evaluation. The physiological states identification is based on the correlation product between biochemical signals. The efficiency of the method has been tested in a fed-batch fermentation having the goal to increase the biomass production.

View all citing articles on Scopus

View full text