Enhancing regression models for complex systems using evolutionary techniques for feature engineering

This work proposes an automatic methodology for modeling complex systems. Our methodology is based on the combination of Grammatical Evolution and classical regression to obtain an optimal set of features that take part of a linear and convex model. This technique provides both Feature Engineering and Symbolic Regression in order to infer accurate models with no effort or designer's expertise requirements. As advanced Cloud services are becoming mainstream, the contribution of data centers in the overall power consumption of modern cities is growing dramatically. These facilities consume from 10 to 100 times more power per square foot than typical office buildings. Modeling the power consumption for these infrastructures is crucial to anticipate the effects of aggressive optimization policies, but accurate and fast power modeling is a complex challenge for high-end servers not yet satisfied by analytical approaches. For this case study, our methodology minimizes error in power prediction. This work has been tested using real Cloud applications resulting on an average error in power estimation of 3.98%. Our work improves the possibilities of deriving Cloud energy efficient policies in Cloud data centers being applicable to other computing environments with similar characteristics.

Engineering Computational Emotion - A Reference Model for Emotion in Artificial Systems

Emotion is generally argued to be an influence on the behavior of life systems, largely concerning flexibility and adaptivity. The way in which life systems acts in response to a particular situations of the environment, has revealed the decisive and crucial importance of this feature in the success of behaviors. And this source of inspiration has influenced the way of thinking artificial systems. During the last decades, artificial systems have undergone such an evolution that each day more are integrated in our daily life. They have become greater in complexity, and the subsequent effects are related to an increased demand of systems that ensure resilience, robustness, availability, security or safety among others. All of them questions that raise quite a fundamental challenges in control design. This thesis has been developed under the framework of the Autonomous System project, a.k.a the ASys-Project. Short-term objectives of immediate application are focused on to design improved systems, and the approaching of intelligence in control strategies. Besides this, long-term objectives underlying ASys-Project concentrate on high order capabilities such as cognition, awareness and autonomy. This thesis is placed within the general fields of Engineery and Emotion science, and provides a theoretical foundation for engineering and designing computational emotion for artificial systems. The starting question that has grounded this thesis aims the problem of emotion--based autonomy. And how to feedback systems with valuable meaning has conformed the general objective. Both the starting question and the general objective, have underlaid the study of emotion, the influence on systems behavior, the key foundations that justify this feature in life systems, how emotion is integrated within the normal operation, and how this entire problem of emotion can be explained in artificial systems. By assuming essential differences concerning structure, purpose and operation between life and artificial systems, the essential motivation has been the exploration of what emotion solves in nature to afterwards analyze analogies for man--made systems. This work provides a reference model in which a collection of entities, relationships, models, functions and informational artifacts, are all interacting to provide the system with non-explicit knowledge under the form of emotion-like relevances. This solution aims to provide a reference model under which to design solutions for emotional operation, but related to the real needs of artificial systems. The proposal consists of a multi-purpose architecture that implement two broad modules in order to attend: (a) the range of processes related to the environment affectation, and (b) the range or processes related to the emotion perception-like and the higher levels of reasoning. This has required an intense and critical analysis beyond the state of the art around the most relevant theories of emotion and technical systems, in order to obtain the required support for those foundations that sustain each model. The problem has been interpreted and is described on the basis of AGSys, an agent assumed with the minimum rationality as to provide the capability to perform emotional assessment. AGSys is a conceptualization of a Model-based Cognitive agent that embodies an inner agent ESys, the responsible of performing the emotional operation inside of AGSys. The solution consists of multiple computational modules working federated, and aimed at conforming a mutual feedback loop between AGSys and ESys. Throughout this solution, the environment and the effects that might influence over the system are described as different problems. While AGSys operates as a common system within the external environment, ESys is designed to operate within a conceptualized inner environment. And this inner environment is built on the basis of those relevances that might occur inside of AGSys in the interaction with the external environment. This allows for a high-quality separate reasoning concerning mission goals defined in AGSys, and emotional goals defined in ESys. This way, it is provided a possible path for high-level reasoning under the influence of goals congruence. High-level reasoning model uses knowledge about emotional goals stability, letting this way new directions in which mission goals might be assessed under the situational state of this stability. This high-level reasoning is grounded by the work of MEP, a model of emotion perception that is thought as an analogy of a well-known theory in emotion science. The work of this model is described under the operation of a recursive-like process labeled as R-Loop, together with a system of emotional goals that are assumed as individual agents. This way, AGSys integrates knowledge that concerns the relation between a perceived object, and the effect which this perception induces on the situational state of the emotional goals. This knowledge enables a high-order system of information that provides the sustain for a high-level reasoning. The extent to which this reasoning might be approached is just delineated and assumed as future work. This thesis has been studied beyond a long range of fields of knowledge. This knowledge can be structured into two main objectives: (a) the fields of psychology, cognitive science, neurology and biological sciences in order to obtain understanding concerning the problem of the emotional phenomena, and (b) a large amount of computer science branches such as Autonomic Computing (AC), Self-adaptive software, Self-X systems, Model Integrated Computing (MIC) or the paradigm of models@runtime among others, in order to obtain knowledge about tools for designing each part of the solution. The final approach has been mainly performed on the basis of the entire acquired knowledge, and described under the fields of Artificial Intelligence, Model-Based Systems (MBS), and additional mathematical formalizations to provide punctual understanding in those cases that it has been required. This approach describes a reference model to feedback systems with valuable meaning, allowing for reasoning with regard to (a) the relationship between the environment and the relevance of the effects on the system, and (b) dynamical evaluations concerning the inner situational state of the system as a result of those effects. And this reasoning provides a framework of distinguishable states of AGSys derived from its own circumstances, that can be assumed as artificial emotion.

Some recent applications of non-linear models to the engineering practice

The Department of Structural Analysis of the University of Santander has been for a longtime involved in the solution of the country´s practical engineering problems. Some of these have required the use of non-conventional methods of analysis, in order to achieve adequate engineering answers. As an example of the increasing application of non-linear computer codes in the nowadays engineering practice, some cases will be briefly presented. In each case, only the main features of the problem involved and the solution used to solve it will be shown

Expressed protein ligation: A general method for protein engineering

A protein semisynthesis method—expressed protein ligation—is described that involves the chemoselective addition of a peptide to a recombinant protein. This method was used to ligate a phosphotyrosine peptide to the C terminus of the protein tyrosine kinase C-terminal Src kinase (Csk). By intercepting a thioester generated in the recombinant protein with an N-terminal cysteine containing synthetic peptide, near quantitative chemical ligation of the peptide to the protein was achieved. The semisynthetic tail-phosphorylated Csk showed evidence of an intramolecular phosphotyrosine-Src homology 2 interaction and an unexpected increase in catalytic phosphoryl transfer efficiency toward a physiologically relevant substrate compared with the non-tail-phosphorylated control. This work illustrates that expressed protein ligation is a simple and powerful new method in protein engineering to introduce sequences of unnatural amino acids, posttranslational modifications, and biophysical probes into proteins of any size.

Engineering a tRNA and aminoacyl-tRNA synthetase for the site-specific incorporation of unnatural amino acids into proteins in vivo

In an effort to expand the scope of protein mutagenesis, we have completed the first steps toward a general method to allow the site-specific incorporation of unnatural amino acids into proteins in vivo. Our approach involves the generation of an “orthogonal” suppressor tRNA that is uniquely acylated in Escherichia coli by an engineered aminoacyl-tRNA synthetase with the desired unnatural amino acid. To this end, eight mutations were introduced into tRNA2Gln based on an analysis of the x-ray crystal structure of the glutaminyl-tRNA aminoacyl synthetase (GlnRS)–tRNA2Gln complex and on previous biochemical data. The resulting tRNA satisfies the minimal requirements for the delivery of an unnatural amino acid: it is not acylated by any endogenous E. coli aminoacyl-tRNA synthetase including GlnRS, and it functions efficiently in protein translation. Repeated rounds of DNA shuffling and oligonucleotide-directed mutagenesis followed by genetic selection resulted in mutant GlnRS enzymes that efficiently acylate the engineered tRNA with glutamine in vitro. The mutant GlnRS and engineered tRNA also constitute a functional synthetase–tRNA pair in vivo. The nature of the GlnRS mutations, which occur both at the protein–tRNA interface and at sites further away, is discussed.

Rational engineering of a miniprotein that reproduces the core of the CD4 site interacting with HIV-1 envelope glycoprotein

Protein–protein interacting surfaces are usually large and intricate, making the rational design of small mimetics of these interfaces a daunting problem. On the basis of a structural similarity between the CDR2-like loop of CD4 and the β-hairpin region of a short scorpion toxin, scyllatoxin, we transferred the side chains of nine residues of CD4, central in the binding to HIV-1 envelope glycoprotein (gp120), to a structurally homologous region of the scorpion toxin scaffold. In competition experiments, the resulting 27-amino acid miniprotein inhibited binding of CD4 to gp120 with a 40 μM IC50. Structural analysis by NMR showed that both the backbone of the chimeric β-hairpin and the introduced side chains adopted conformations similar to those of the parent CD4. Systematic single mutations suggested that most CD4 residues from the CDR2-like loop were reproduced in the miniprotein, including the critical Phe-43. The structural and functional analysis performed suggested five additional mutations that, once incorporated in the miniprotein, increased its affinity for gp120 by 100-fold to an IC50 of 0.1–1.0 μM, depending on viral strains. The resulting mini-CD4 inhibited infection of CD4+ cells by different virus isolates. Thus, core regions of large protein–protein interfaces can be reproduced in miniprotein scaffolds, offering possibilities for the development of inhibitors of protein–protein interactions that may represent useful tools in biology and in drug discovery.

Engineering the largest RNA virus genome as an infectious bacterial artificial chromosome

The construction of cDNA clones encoding large-size RNA molecules of biological interest, like coronavirus genomes, which are among the largest mature RNA molecules known to biology, has been hampered by the instability of those cDNAs in bacteria. Herein, we show that the application of two strategies, cloning of the cDNAs into a bacterial artificial chromosome and nuclear expression of RNAs that are typically produced within the cytoplasm, is useful for the engineering of large RNA molecules. A cDNA encoding an infectious coronavirus RNA genome has been cloned as a bacterial artificial chromosome. The rescued coronavirus conserved all of the genetic markers introduced throughout the sequence and showed a standard mRNA pattern and the antigenic characteristics expected for the synthetic virus. The cDNA was transcribed within the nucleus, and the RNA translocated to the cytoplasm. Interestingly, the recovered virus had essentially the same sequence as the original one, and no splicing was observed. The cDNA was derived from an attenuated isolate that replicates exclusively in the respiratory tract of swine. During the engineering of the infectious cDNA, the spike gene of the virus was replaced by the spike gene of an enteric isolate. The synthetic virus replicated abundantly in the enteric tract and was fully virulent, demonstrating that the tropism and virulence of the recovered coronavirus can be modified. This demonstration opens up the possibility of employing this infectious cDNA as a vector for vaccine development in human, porcine, canine, and feline species susceptible to group 1 coronaviruses.

Tissue engineering of cartilage in space

Tissue engineering of cartilage, i.e., the in vitro cultivation of cartilage cells on synthetic polymer scaffolds, was studied on the Mir Space Station and on Earth. Specifically, three-dimensional cell-polymer constructs consisting of bovine articular chondrocytes and polyglycolic acid scaffolds were grown in rotating bioreactors, first for 3 months on Earth and then for an additional 4 months on either Mir (10−4–10−6 g) or Earth (1 g). This mission provided a unique opportunity to study the feasibility of long-term cell culture flight experiments and to assess the effects of spaceflight on the growth and function of a model musculoskeletal tissue. Both environments yielded cartilaginous constructs, each weighing between 0.3 and 0.4 g and consisting of viable, differentiated cells that synthesized proteoglycan and type II collagen. Compared with the Earth group, Mir-grown constructs were more spherical, smaller, and mechanically inferior. The same bioreactor system can be used for a variety of controlled microgravity studies of cartilage and other tissues. These results may have implications for human spaceflight, e.g., a Mars mission, and clinical medicine, e.g., improved understanding of the effects of pseudo-weightlessness in prolonged immobilization, hydrotherapy, and intrauterine development.

Propagation of an attenuated virus by design: engineering a novel receptor for a noninfectious foot-and-mouth disease virus.

To gain entry into cells, viruses utilize a variety of different cell-surface molecules. Foot-and-mouth disease virus (FMDV) binds to cell-surface integrin molecules via an arginine-glycine-aspartic acid (RGD) sequence in capsid protein VP1. Binding to this particular cell-surface molecule influences FMDV tropism, and virus/receptor interactions appear to be responsible, in part, for selection of antigenic variants. To study early events of virus-cell interaction, we engineered an alternative and novel receptor for FMDV. Specifically, we generated a new receptor by fusing a virus-binding, single-chain antibody (scAb) to intracellular adhesion molecule 1 (ICAM1). Cells that are normally not susceptible to FMDV infection became susceptible after being transfected with DNA encoding the scAb/ICAM1 protein. An escape mutant (B2PD.3), derived with the mAb used to generate the genetically engineered receptor, was restricted for growth on the scAb/ICAM1 cells, but a variant of B2PD.3 selected by propagation on scAb/ICAM1 cells grew well on these cells. This variant partially regained wild-type sequence in the epitope recognized by the mAb and also regained the ability to be neutralize by the mAb. Moreover, RGD-deleted virions that are noninfectious in animals and other cell types grew to high titers and were able to form plaques on scAb/ ICAM1 cells. These studies demonstrate the first production of a totally synthetic cell-surface receptor for a virus. This novel approach will be useful for studying virus reception and for the development of safer vaccines against viral pathogens of animals and humans.

Genetic engineering of vein grafts resistant to atherosclerosis.

Previously, researchers have speculated that genetic engineering can improve the long-term function of vascular grafts which are prone to atherosclerosis and occlusion. In this study, we demonstrated that an intraoperative gene therapy approach using antisense oligodeoxynucleotide blockage of medial smooth muscle cell proliferation can prevent the accelerated atherosclerosis that is responsible for autologous vein graft failure. Selective blockade of the expression of genes for two cell cycle regulatory proteins, proliferating cell nuclear antigen and cell division cycle 2 kinase, was achieved in the smooth muscle cells of rabbit jugular veins grafted into the carotid arteries. This alteration of gene expression successfully redirected vein graft biology away from neointimal hyperplasia and toward medial hypertrophy, yielding conduits that more closely resembled normal arteries. More importantly, these genetically engineered grafts proved resistant to diet-induced atherosclerosis. These findings establish the feasibility of developing genetically engineered bioprostheses that are resistant to failure and better suited to the long-term treatment of occlusive vascular disease.

Particle engineering for the production of respirable dry powder formulations

Hyaluronan (HA) plays an important role in lung pathophysiology. For this reason it has attracted great attention both as active ingredient and as excipient in treating lung diseases by direct pulmonary HA administration. The aim was the production of highly respirable and flowable HA powders either as a potential carrier for drug delivery or for being delivered directly by inhalation. Engineered sodium hyaluronate powders were produced by spray-drying technique. All the spray-dried powders were characterised in terms of particle size distribution, drug content, morphology and in vitro respirability. HA was successfully formulated with salbutamol sulphate in combination with leucine and highlighted remarkable aerodynamic performance (emitted dose equal to 83 % and FPF % equal to 97.1%). Moreover, HA colloidal solutions were designed and they were spray-dried. In order to improve particle aerodynamic characteristics, different types of excipients were investigated. In particular, stearylamine (5% w/w) allowed to obtain the best performance throughout the experimental set. Finally, in vitro biocompatibility was carried out by MTT assay and High Content Analysis for selected dry powder formulations and starting materials. The assays demonstrated the same outcome by confirming the HA biocompatibility and by producing the same rank of toxicity for the surfactants. The general conclusion of the project is that formulation containing HA and stearyl alcohol represents the best performing formulation.

Towards a reverse engineering approach for guiding user in applying data mining

Comunicación presentada en las XVI Jornadas de Ingeniería del Software y Bases de Datos, JISBD 2011, A Coruña, 5-7 septiembre 2011.

Image processing for safety assessment in Civil Engineering

Behaviour analysis of construction safety systems is of fundamental importance to avoid accidental injuries. Traditionally, measurements of dynamic actions in Civil Engineering have been done through accelerometers, but high-speed cameras and image processing techniques can play an important role in this area. Here, we propose using morphological image filtering and Hough transform on high-speed video sequence as tools for dynamic measurements on that field. The presented method is applied to obtain the trajectory and acceleration of a cylindrical ballast falling from a building and trapped by a thread net. Results show that safety recommendations given in construction codes can be potentially dangerous for workers.