Machine learning methods for prediction of disulphide bonding states of cysteine residues in proteins

The goal of this thesis work is to develop a computational method based on machine learning techniques for predicting disulfide-bonding states of cysteine residues in proteins, which is a sub-problem of a bigger and yet unsolved problem of protein structure prediction. Improvement in the prediction of disulfide bonding states of cysteine residues will help in putting a constraint in the three dimensional (3D) space of the respective protein structure, and thus will eventually help in the prediction of 3D structure of proteins. Results of this work will have direct implications in site-directed mutational studies of proteins, proteins engineering and the problem of protein folding. We have used a combination of Artificial Neural Network (ANN) and Hidden Markov Model (HMM), the so-called Hidden Neural Network (HNN) as a machine learning technique to develop our prediction method. By using different global and local features of proteins (specifically profiles, parity of cysteine residues, average cysteine conservation, correlated mutation, sub-cellular localization, and signal peptide) as inputs and considering Eukaryotes and Prokaryotes separately we have reached to a remarkable accuracy of 94% on cysteine basis for both Eukaryotic and Prokaryotic datasets, and an accuracy of 90% and 93% on protein basis for Eukaryotic dataset and Prokaryotic dataset respectively. These accuracies are best so far ever reached by any existing prediction methods, and thus our prediction method has outperformed all the previously developed approaches and therefore is more reliable. Most interesting part of this thesis work is the differences in the prediction performances of Eukaryotes and Prokaryotes at the basic level of input coding when ‘profile’ information was given as input to our prediction method. And one of the reasons for this we discover is the difference in the amino acid composition of the local environment of bonded and free cysteine residues in Eukaryotes and Prokaryotes. Eukaryotic bonded cysteine examples have a ‘symmetric-cysteine-rich’ environment, where as Prokaryotic bonded examples lack it.

Pharmaceutical residues in aquatic systems: mode of action and effects on mussel physiology

Pharmaceutical residues contaminate aquatic ecosystems as a result of their widespread human and veterinary usage. Since continuously released and not efficiently removed, certain pharmaceuticals exhibit pseudo-persistence thus generating concerns for the health of aquatic wildlife. This work aimed at assessing on mussels Mytilus galloprovincialis, under laboratory conditions, the effects of three pharmaceuticals, carbamazepine (antiepileptic), propranolol (β-blocker) and oxytetracycline (antibiotic), to evaluate if the human-based mode of action of these molecules is conserved in invertebrates. Furthermore, in the framework of the European MEECE Programme, mussels were exposed to oxytetracycline and copper at increasing temperatures, simulating variations due to climate changes. The effects of these compounds were assessed evaluating a battery of biomarkers, the expression of HSP70 proteins and changes in cAMP-related parameters. A decrease in lysosomal membrane stability, induction of oxidative stress, alterations of cAMP-dependent pathway and the induction of defense mechanisms were observed indicating the development of a stress syndrome, and a worsening in mussels health status. Data obtained in MEECE Programme confirmed that the toxicity of substances can be enhanced following changes in temperature. The alterations observed were obtained after exposure to pharmaceuticals at concentrations sometimes lower than those detected in the aquatic environment. Hence, further research is advisable regarding subtle effects of pharmaceuticals on non-target organisms. Furthermore, results obtained during a research stay in the laboratories of Cádiz University (Spain) are presented. The project aimed at measuring possible effects of polluted sediments in Algeciras Bay (Spain) and in Cádiz Bay, by assessing different physiological parameters in caged crabs Carcinus maenas and clams Ruditapes decussatus exposed in situ for 28 days. The neutral red retention assay was adapted to these species and proved to be a sensitive screening tool for the assessment of sediment quality.

Study of innovative methods of control in the cereal productive chain for the production of beer and spirits

Researches performed during the PhD course intended to assess innovative applications of near-infrared spectroscopy in reflectance (NIR) in the production chain of beer. The purpose is to measure by NIR the "malting quality" (MQ) parameter of barley, to monitor the malting process and to know if a certain type of barley is suitable for the production of beer and spirits. Moreover, NIR will be applied to monitor the brewing process. First of all, it was possible to check the quality of the raw materials like barley, maize and barley malt using a rapid, non-destructive and reliable method, with a low error of prediction. The more interesting result obtained at this level was that the repeatability of the NIR calibration models developed was comparable with the one of the reference method. Moreover, about malt, new kinds of validation were used in order to estimate the real predictive power of the proposed calibration models and to understand the long-term effects. Furthermore, the precision of all the calibration models developed for malt evaluation was estimated and statistically compared with the reference methods, with good results. Then, new calibration models were developed for monitoring the malting process, measuring the moisture content and other malt quality parameters during germination. Moreover it was possible to obtain by NIR an estimate of the "malting quality" (MQ) of barley and to predict whether if its germination will be rapid and uniform and if a certain type of barley is suitable for the production of beer and spirits. Finally, the NIR technique was applied to monitor the brewing process, using correlations between NIR spectra of beer and analytical parameters, and to assess beer quality. These innovative results are potentially very useful for the actors involved in the beer production chain, especially the calibration models suitable for the control of the malting process and for the assessment of the “malting quality” of barley, which need to be deepened in future studies.