Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

http://dbkgroup.org/Papers/davey_kell_micrev96.pdf

Flow cytometry and cell sorting of heterogeneous microbial populations: the importance of single-cell analyses.

Starting with a brief history of solid-state fermentation (SSF), major aspects of SSF are reviewed, which include factors affecting SSF, biomass, fermentors, modeling, industrial microbial enzymes, organic acids, secondary metabolites, and bioremediation. Physico-chemical and environmental factors such as inoculum type, moisture and water activity, pH, temperature, substrate, particle size, aeration and agitation, nutritional factors, and oxygen and carbon dioxide affecting SSF are reviewed. The advantages of SSF over Submerged Fermentation (SmF) are indicated, and the different types of fermentors used in SSF described. The economic feasibilities of adopting SSF technology in the commercial production of industrial enzymes such as amylases, cellulases, xylanase, proteases, phytases, lipases, etc., organic acids such as citric acid and lactic acid, and secondary metabolites such as gibberellic acid, ergot alkaloids, and antibiotics such as penicillin, cyclosporin, cephamycin and tetracyclines are highlighted. The relevance of applying SSF technology in the production of mycotoxins, biofuels, and biocontrol agents is discussed, and the need for adopting SSF technology in bioremediation of toxic compounds, biological detoxication of agro-industrial residues, and biotransformation of agro-products and residues is emphasized.

Solid-State Fermentation Systems—An Overview

In microbiology the terms 'viability' and 'culturability' are often equated. However, in recent years the apparently self-contradictory expression 'viable-but-nonculturable' ('VBNC') has been applied to cells with various and often poorly defined physiological attributes but which, nonetheless, could not be cultured by methods normally appropriate to the organism concerned. These attributes include apparent cell integrity, the possession of some form of measurable cellular activity and the apparent capacity to regain culturability. We review the evidence relating to putative VBNC cells and stress our view that most of the reports claiming a return to culturability have failed to exclude the regrowth of a limited number of cells which had never lost culturability. We argue that failure to differentiate clearly between use of the terms 'viability' and 'culturability' in an operational versus a conceptual sense is fuelling the current debate, and conclude with a number of proposals that are designed to help clarify the major issues involved. In particular, we suggest an alternative operational terminology that replaces 'VBNC' with expressions that are internally consistent.

/pdf/viability-and-activity-in-readily-culturable-bacteria-a-13n1hqbg6g.pdf

Viability and activity in readily culturable bacteria: a review and discussion of the practical issues

http://www.ejbiotechnology.info/content/vol1/issue3/full/9/9.PDF

General and microbiological aspects of solid substrate fermentation

The dielectric properties of biological cells at radiofrequencies : Applications in biotechnology

We review physical approaches to the problem of devising a real-time biomass probe. Direct measurement of the dielectric permittivity of cell suspensions at radio frequencies provides one possible solution to this problem.

http://dbkgroup.org/Papers/kell_biomass_trac90.pdf

Real-time monitoring of cellular biomass: Methods and applications

The dielectric properties of human erythrocytes (red blood cells) suspended in whole blood and in isotonic media at various volume fractions (haematocrits) have been studied in the frequency range 0.2-10 MHz, in which the so-called beta-dispersion due to the Maxwell-Wagner effect is known to occur. The capacitance and conductance at 25 degrees C were measured by an instrument interfaced to a computer. The rectangular sample cavity (1 ml volume) contained four pure gold electrode pins, and the sample could be circulated by a roller pump. The frequency-dependence of the permittivity and conductivity were fitted by non-linear least squares regression. Corrections were applied for non-linearity in the dielectric increment at high haematocrit, and for electrode polarisation when diluting the blood in saline. Data were interpreted in terms of a simple equivalent resistor-capacitor circuit. From the measured haematological values the specific membrane capacitance (Cm) and the conductivities internal and external to the cells (sigma i' and sigma o' respectively) were estimated. The conductivities behaved in a predictable manner with a mean of 0.458 S.m-1 (s.d. +/- 0.044) for sigma i', whereas the value of Cm (and indeed the actual capacitance of the suspension) was dependent on the amount of plasma present. Hence, in stationary normal (anticoagulated) whole blood samples, Cm was as high as 2.98 mu F.cm-2 (s.d. +/- 0.40), in contrast to about 0.9 mu F.cm-2 in blood diluted more than two-fold (to less than 20% hct) in isotonic media. The high value remained when the diluent was plasma.(ABSTRACT TRUNCATED AT 250 WORDS)

/pdf/dielectric-properties-of-human-blood-and-erythrocytes-at-4nbvywx18g.pdf

Dielectric properties of human blood and erythrocytes at radio frequencies (0.2–10 MHz); dependence on cell volume fraction and medium composition

Introduction All else being equal, the productivity of a biological process is determined by the quantity of biomass present. There is therefore a major requirement for the accurate measurement and control of the biomass within fermentors, at both laboratory and industrial scales. Presently the range of sensors available that can be used in situ and reliably for the monitoring and regulation of biotechnological processes in general is rather limited. These sensors normally rely upon physical (e.g. optical, mechanical and electrical) or chemical variables (e.g. pH and concentration) rather than biological ones per se (Sarra et al., 1996; Pons, 1991). However only physical methods allow the on-line, real-time estimation of biomass (Harris and Kell, 1985). As well as physical methods, any easily determinable chemical that is produced or consumed by cells at an essentially constant rate during cell growth may also be used to assess biomass, e.g. carbon dioxide evolution and oxygen consumption. In these indirect methods biomass is then calculated based upon mass balances, stoichiometric relationships or empirical constants. However, this type of approach has the great disadvantage that it does not generally discriminate between biomass and necromass (Kell et al., 1990). Even if biomass was easily measurable there is still the question of what is biologically relevant information for fermentation control and how can one define and quantify it (e.g. metabolism, viability, vitality, morphology) (Kell et al., 1987; Kell, 1987a;

/pdf/on-line-real-time-measurements-of-cellular-biomass-using-2cbyezfbzg.pdf

On-line, real time measurements of cellular biomass using dielectric spectroscopy

We describe a continuous culture system related to the turbidostat, but using a feedback system based on biomass estimation from the dielectric permittivity of the cell suspension rather than its optical density. It is shown that this system provides an excellent method of maintaining a constant biomass level within a fermentor. The computer-controlled system was able to effect the essentially continuous registration of growth rate by monitoring the rate of medium addition via the time-dependent activity of the pump. At some biomass setpoints for aerobically grown cultures of baker's yeast substantial time-dependent fluctuations in the growth rate of the culture were thereby observed. At some biomass setpoints, however, or under anaerobic conditions, or when using a non-Crabtree yeast, the growth rate was constant, indicating that the fluctuations were inherent to the biological system and not simply a property of the fermentor and control system. A variety of time series analyses (Fourier transformations, Hurst and Lyapunov exponents, the determination of embedding dimension, and non-linear time series predictions based on the methodology of Sugihara and May) were used to demonstrate, for the first time, that as well as stochastic and periodic components these fluctuations exhibited deterministic chaos. ‘Trivial predictors’ were unable to give accurate predictions of the growth rate in these cultures. The growth rate fluctuations were studied further by means of offline measurements of changes in percentage viability, bud count, and in the external ethanol and glucose concentrations; these data and other evidence suggested that the growth rate fluctuations were closely linked to the primary respiro-fermentative metabolism of this organism. The identification of chaotic growth rates in cell cultures suggests that there may be novel methods for controlling the growth of such cultures.

http://dbkgroup.org/Papers/davey_chaos_biosystems96.pdf

Christopher L. Davey

Papers

The dielectric properties of biological cells at radiofrequencies : Applications in biotechnology

Real-time monitoring of cellular biomass: Methods and applications

Dielectric properties of human blood and erythrocytes at radio frequencies (0.2–10 MHz); dependence on cell volume fraction and medium composition

On-line, real time measurements of cellular biomass using dielectric spectroscopy

Oscillatory, stochastic and chaotic growth rate fluctuations in permittistatically controlled yeast cultures