A DNA-based biomolecular string processing scheme demonstrated by Adleman has attracted wide attention. While it is not known to what degree the scheme can scale up, it nevertheless introduces a new and interesting concept which seems so far to have been overlooked. The key point is that the Adleman scheme involves building specific hardware for a single problem instance. This opens a design degree of freedom that is not limited to biomolecular architectures.

Natural biomolecular systems process information in a radically different manner than programmable machines. Conformational interactions, the basis of specificity and self-assembly, are of key importance. A gedanken device is presented that illustrates how the fusion of information through conformational self-organization can serve to enhance pattern processing at the cellular level. The device is used to highlight general features of biomolecular information processing. We briefly outline a simulation system designed to address the manner in which conformational processing interacts with kinetic and higher level structural dynamics in complex biochemical networks. Virtual models that capture features of biomolecular information processing can in some instances have artificial intelligence value in their own right and should serve as design tools for future computers built from real molecules.

Evolve IV is designed to explore the effects of environmental uncertainty on niche proliferation and the evolution of interspecific interactions. Preliminary experiments with the model suggest that autotrophicheterotrophic specializations emerge.

The capabilities built into a processing network control the manner in which it generalizes from a training set and therefore how it groups environmental patterns. These capabilities are developed through learning, ultimately evolutionary learning, and therefore have an objective basis in so far as the grouping tendencies afford a selective advantage. But for this development to occur it is necessary that the processing network in fact be able to evolve grouping tendencies that reflect selective pressures. The extent to which this is possible depends on how wide a variety of grouping dynamics the processing network can support (its dynamic richness) and on whether its structure-function gradualism (evolutionary friendliness) is sufficient to provide access to these grouping responses through a variation-selection process. We describe a “softened” cellular automaton model that illustrates how different grouping responses can be evolved in cases simple enough to examine the entire test set.

High dimensionality and interactional complexity, appropriately introduced, can enhance the evolvability of a pattern processing network. We describe a processor, referred to as the cytomatrix module, that can be used to investigate the requisite conditions for such enhancement. The processor is characterized by multiplicity of component types, graded interactions among components, separation of signal integration dynamics from the readout mechanisms that interpret these dynamics, and multiplicity of parameters open to evolution (including component connectivity). The adaptation procedure is mediated by a multiparameter variation–selection algorithm that acts on the various parameters in an alternating (i.e., phasic) manner. Experiments with both structured and unstructured learning tasks, as well as with difficult parity problems, demonstrate that opening more parameters to evolution increases the flexibility exhibited by the processor in response to evolutionary pressure, essentially by loosening the coupling between the local and global aspects of the response. The cytomatrix processor can be thought of as a highly abstracted representation of signal integration within single neurons; alternatively, it can be viewed as a collection of cells in a multicellular organization.

Loosely bound electrons in proteins allow for interference effects that focus thermal energy on selected degrees of nuclear freedom. The importance for enzyme speed and specificity is suggested by restrictions that conformational motions would have to comply with if all electrons were required to behave in accordance with the quantum mechanical adiabatic principle. The electronic superpositions lead as well as follow the nuclear motions.

Issues addressed in H.H. Pattee's origin of life laboratory in the 1960s and their connection to the physics-evolution-language problematic are indicated. The problem of quantum measurement played a central role. The problem is herein examined in the light of the fluctuon model; in particular, as the model applies to gravity. The main conclusion is that measurement and motion are a unitary process. All accelerations are accompanied by a cycle involving the annihilation and creation of superpositions. Gravitational collapse is also a cyclic process in the fluctuon model. By a suitable transformation, it can be seen that interactions underlying superpositional collapse are the same as those operative in gravitational collapse. Implications for the origin of cellular life and the development of symbolic systems are considered.

Cells and organisms are natural molecular computers. The problem domains effectively addressed by these systems are complementary to, and in fundamental respects far exceed, the domains addressable by current computing devices. The vast majority of information processing problems do not have sufficiently compact formal specifications to fall within the reach of programmable machines. Our working hypothesis is that the unique properties of molecular materials are the key to extending information processing technology beyond the narrow limits of formal computing.

The conformational dynamics of enzymes is a computational resource that fuses milieu signals in a nonlinear fashion. Response surface methodology can be used to elicit computational functionality from enzyme dynamics. We constructed a tabletop prototype to implement enzymatic signal processing in a device context and employed it in conjunction with malate dehydrogenase to perform the linearly inseparable exclusive-or operation. This shows that proteins can execute signal processing operations that are more complex than those performed by individual threshold elements. We view the experiments reported, though restricted to the two-variable case, as a stepping stone to computational networks that utilize the precise reproducibility of proteins, and the concomitant reproducibility of their nonlinear dynamics, to implement complex pattern transformations.

The hypernetwork model is a hierarchical architecture that has a representation of the molecular, cellular, and organismic levels of biological organization. It influences flow within each level, and through levels, forming dynamic networks of molecular interactions. With its molecular variation-selection learning algorithm, the hypernetwork is able to solve fairly complex tasks such as the (4-10)-input parity task, and the tic-tac-toe endgame problem, with good results. These performance results illustrate the learning capabilities of this model

Today's computers are built up from a minimal set of standard pattern recognition operations. Logic gates, such as NAND, are common examples. Biomolecular materials offer an alternative approach, both in terms of variety and context sensitivity. Enzymes, the basic switching elements in biological cells, are notable for their ability to discriminate specific molecules in a complex background and to do so in a manner that is sensitive to particular milieu features and indifferent to others. The enzyme, in effect, is a powerful context sensitive pattern recognizer. We describe a tabletop pattern processor that in a rough way can be analogized to a neuron whose input-output behavior is controlled by enzymatic dynamics.

A confluence of factors emanating from computer science, biology, and technology have brought self-organizing approaches back to the fore. Neural networks in particular provide high evolvability platforms for variation-selection search strategies. The neuron doctrine and the fundamental nature of computing come into question. Is a neuron an atom of the brain or is it itself a complex information processing system whose interior molecular dynamics can be elicited and exploited through the evolution process? We argue the latter point of view, illustrating how high evolvability dynamics can be achieved with artificial neuromolecular computer designs and how such designs might in due course be implemented using molecular computing devices. A tabletop enzyme-driven prototype recently implemented in our laboratory is briefly described; it can be thought of as a sort of artificial neuron in which the context sensitivity of enzyme recognition is used to transform injected signal patterns into output activity.

A table-top prototype has been constructed that uses the enzyme malate dehydrogenase to recognize input signal patterns. The device is controlled by the enzyme in response to injection of Mg2+ used as a signaling substance. Output is monitored spectroscopically. If Mg2+ is injected along either of two signal lines (i.e., if the input signal pattern is 10 or 01) the device emits an output of 1. Injection along neither or both lines results in an output of 0. The enzyme in effect is used as a transform that converts the linearly inseparable exclusive- or problem into a linearly separable problem.

A hierarchical architecture for information processing, the hypernetwork model, has recently been implemented. This is a three level model, inspired by biological systems, that includes representation of scale, vertical flow of information, and feedback control. All interactions are based on complementary relationships between molecular subunits. The system is molded to perform desired tasks through a variation-selection algorithm acting on the structure of the molecular subunits. The design of the system, the learning algorithm, and preliminary pattern classification results are presented

Proteins and nucleic acids constitute a vast potential reservoir of pattern recognizers that operate on the basis of shape complementarity. It is possible to construct models of computing in which these shape-based interactions contribute directly to recognition of signal patterns at the device (or cell) level. The input–output transform is molded by variation-selection evolution. Such models provide clues as to the organizational features that enable biomolecular matter to acquire nonevolutionary modes of problem solving through the evolutionary process. The requisite organizations are characterized by a high dimensionality that allows them to simultaneously exhibit aspects of context-sensitivity and context-independence.