As a remarkable class of nanomaterials, research on hydrogels is developing rapidly through the innovation and improvement of polymers, chemical synthesis and methods of preparation, formulations and fabrication process, as well as applications. This overview concerns the analysis of international patents on hydrogels through the global system for filing patent applications which is known as Patent Cooperation Treaty (PCT) and administered by Intellectual Property Organization (WIPO). More specifically, this study presents the state of the art by introducing what has been innovated and patented in relation to hydrogels. As results, a total of 11,757 patent applications related to hydrogels have been filled between 1979 and 2020. The United States leads the patent race in the hydrogel sector. Massachusetts Institute of Technology is one of the top innovators in the hydrogel-based research and development. Based on patent classifications, most patent applications are related to medicinal preparations characterised by special physical form and materials for prostheses or for coating prostheses including structure of prostheses use of preparations for artificial biological organs, as well as materials characterised by their function or physical properties, such as hydrogels or hydrocolloids.

There are a variety of hydrogels commonly used as coatings for medical device applications. The device coating process involving the deposition of hydrogels to the device surface to enhance its performance, particularly, through promotion of osseointegration, haemocompatibility, lubricity and resistance to biofouling. Hydrogels are synthetic matrices made up of a network of hydrophilic polymers that absorb water and/or biological fluids. They can be created from a large number of water-soluble materials including synthetic polymers and biopolymers. The 3D structure of these hydrogels is due to crosslinking which forms an insoluble macromolecular network in the environmental fluid. Research on hydrogels as biocompatible coatings is developing rapidly through the innovation and improvement of raw materials (polymers), chemical synthesis and methods of preparation, as well as formulations and fabrication process. This trend is justified by the several advantages that hydrogels offer for biofabrication and biomedical applications. This work in the form of patentability study presents the state of the art by introducing what has been innovated and patented in relation to hydrogel coatings. Furthermore, a detailed analysis of the patentability of hydrogel applications such as in the coating of medical devices to enhance its clinical performance, have been provided by determining publication years, jurisdictions, inventors, applicants, owners, and classifications.

The contamination of food by microorganisms, their persistence, growth, multiplication, and/or toxin production has emerged as an important public health concern. The demand for consuming fresh and low-processed foods free of chemicals and pathogens is increasing. Despite advances in food safety, annually, more than 9 million persons developed illnesses caused by food contamination (Scallan et al., 2011). In Ecuador, the risk of diseases associated with food contaminations is increasing due to incorrect food manipulation, hygiene, and inappropriate storage conditions (Garzon et al., 2017). Although the vendors are continuously capacitated, no improvement on selling sites was made. The food is continuously sold on streets, near parks, transportation terminals, as a common habit. Along with the excessive use of chemicals for the purpose of preservation, food safety is of concern. To overcome this problem, the application of natural methods for preservation might be a suitable solution. Lactic acid bacteria are producing peptides or small proteins namely bacteriocins which could be the next generation of antimicrobials. Thus, their incorporation in food to prevent poisoning or spoilage has been an area of dynamic research in the last decade (Backialakshmi et al., 2015). Previously, we identified two native bacteriocinogenic strains, Lactobacillus plantarum UTNGt2 and L. plantarum UTNCys5-4, producing peptides with a broad spectrum of antibacterial activity against several foodborne pathogens in vitro (Tenea and Pozo, 2019; Tenea and Guana, 2019). Moreover, the addition of those peptide extracts at the exponential phase of growth of the target bacteria (Staphylococcus aureus ATCC1026) results in a decrease of total cell viability with about 3.2-fold (log CFU/ml) order of magnitude at 6 h of incubation, indicating their bactericidal mode of action. In this study, the possible mechanism of action against Staphylococcus aureus was investigated through a series of cell biology analyses such as membrane permeabilization, cell integrity, and structural changes of the target cells. Altogether, the results demonstrated the effectiveness of peptides produced by native lactic acid bacteria to kill Staphylococcus and further investigation is need it to prove the effect in a food matrix.

1. _{Scallan, E., Hoekstra, R. M., Angulo, F. J., Tauxe, R. V., Widdowson, M. A., Roy, S. L., Jones, J. L., & Griffin, P. M. (2011). Foodborne illness acquired in the United States--major pathogens.Emerging infectious diseases, 17(1), 7–15. https://doi.org/10.3201/eid1701.p11101.}
2. _{Garzón, K., Ortega, C., & Tenea, G. N. (2017). Characterization of Bacteriocin-Producing Lactic Acid Bacteria Isolated from Native Fruits of Ecuadorian Amazon.Polish journal of microbiology, 66(4), 473–481. https://doi.org/10.5604/01.3001.0010.7037}
3. _{Backialakshmi, S., Rn, M., Saranya, A., Ms, J.T., Ar, K., Js, K., & Ramasamy, S. (2015). Biopreservation of Fresh Orange Juice Using Antilisterial Bacteriocins101 and Antilisterial Bacteriocin103 Purified from Leuconostoc mesenteroides.Journal of Food Processing and Technology, 6, 1-5.}
4. _{Tenea, G. N., & Delgado Pozo, T. (2019). Antimicrobial Peptides fromLactobacillus plantarum UTNGt2 Prevent Harmful Bacteria Growth on Fresh Tomatoes. Journal of microbiology and biotechnology, 29(10), 1553–1560. https://doi.org/10.4014/jmb.1904.04063}
5. _{Tenea, G. N., & Guana, J. M. (2019). Inhibitory substances produced by native Lactobacillus plantarum UTNCys5-4 control microbial population growth in meat. Journal of Food Quality, a9516981. https://doi.org/10.1155/2019/9516981.}
6. _{Wang, X., Teng, D., Mao, R., Yang, N., Hao, Y., & Wang, J. (2016). Combined Systems Approaches Reveal a Multistage Mode of Action of a Marine Antimicrobial Peptide against Pathogenic Escherichia coli and Its Protective Effect against Bacterial Peritonitis and Endotoxemia.Antimicrobial agents and chemotherapy, 61(1), e01056-16. https://doi.org/10.1128/AAC.01056-16.}

Osteoarthritis of the intervertebral disc of the lumbosacral spine accounts for approximately 15% of all absences from work. It is also the most common non-traumatic cause of disability in patients up to 45 years of age. The aim of the study was to evaluate the concentration of alanine aminotransferase, aspartate aminotransferase and urea, in a group of 60 patients with osteoarthritis of the lumbosacral spine qualified for microdiscectomy (S group) compared to 60 healthy volunteers (C group). The patients declared that due to pain, they would take non-steroidal painkillers for an average of 2 years. 40% of patients used drugs in the maximum allowable doses. Patients qualified for surgery had a score of 5-7 on the Pffirrman severity of degenerative changes. Statistical analysis showed significant differences in the concentration of the assessed markers: alanine aminotransferase: S vs. C 35.7 U/l vs 9.4U/l; p = 0,000297 ; aspartate aminotransferase S vs. C 26.3 U/l vs. 11.6U/l; p = 0,000060; urea 5.62 mmol/L vs 2.96 mmol/L; p= 0.000006. Reported higher levels of liver enzymes and urea in patients with osteoarthritis of the spine may be due to several reasons. First, it may be the result of long-term use of high doses of non-steroidal pain medications. Secondly, it may indirectly indicate microdamages of skeletal muscles and their hypoxia. Of course, the values of the assessed parameters are not significantly increased, but they indicate the need to limit pharmacotherapy with analgesics in favor of physiotherapy or prior neurosurgery.

The emergence of Multidrug-Resistant (MDR) strains promotes the improvement of Antibacterial Drugs (AD). Some nanoparticles (NP) may be AD carriers, but some have antibacterial activity per se. This opens a window of opportunity for the design of Dual Antibacterial Drug-Nanoparticle (DADNP) systems. DADNP discovery is a slow process due to the high number of combinations of NP vs. AD compounds, assays, etc. Artificial Intelligence/Machine Learning (AI/ML) algorithms that anticipate which potential DADNP systems should be shortlisted for assay may speed up the process. Despite this, the low amount of DADNP activity indicates that AI/ML analysis is tough. To solve this problem in an additive manner, the IFPTML = Information Fusion (IF) + Perturbation-Theory (PT) + Machine Learning (ML) technique was applied. Two datasets were combined (>165000 ChEMBL AD experiments with 300 NP assays) against multiple bacteria species. Next, all vectors of AD and NP properties and experimental conditions (Ddk, Dnk, cdj, and cnj) were zipped into a few input PT Operators (PTOs). IFPTML-LDA models show an Accuracy ≈ 89%, Specificity ≈ 90% and Sensibility ≈ 74% in the training/validation series. The IFPTML models may become a useful tool in the design of DADNP systems for antibacterial therapy against multidrug-resistant microbial pathogens.

Metabolic Reaction Networks (MRNs) are complex networks produced by thousands of chemical reactions or transformations (links) of metabolites (nodes) in a live organism. An essential goal of chemical biology is to test the connectivity (structure) of these complex MRNs models presented for new microorganisms with promising features. In theory, we can undertake hands-on testing (Manual Curation). However, due to the large number of possible combinations of node pairs, this is a difficult operation (possible metabolic reactions). We combined Combinatorial, Perturbation Theory, and Machine Learning approaches in this study to find a CPTML model for MRNs >40 organisms compiled by Barabasis' group. First, we used a novel type of node index termed Markov linear indices fk to quantify the local structure of a very large collection of nodes in each MRN. Next, for over 150 000 MRN query and reference node combinations, we computed CPT operators. Finally, we fed these CPT operators into several ML algorithms. The CPTML linear model obtained using the LDA algorithm is capable of distinguishing nodes (metabolites) with correct reaction assignment from nodes with incorrect reaction assignment with accuracy, specificity, and sensitivity values ranging from 85 to 100 % in both the training and external validation data series. Meanwhile, the top three non-linear models with more than 97.5 % accuracy were found to be PTML models based on Bayesian networks, J48-Decision Tree, and Random Forest algorithms. The new work sets the door for the investigation of MRNs from various organisms using PTML models. Finally, the new CPTML could be a useful tool for determining the structure of MRNs in new species in biotechnology.

The rise of new infectious diseases, combined with an increase in antibiotic resistance among bacterial pathogens, poses a significant health danger to humans. This ever-increasing threat of bacterial resistance necessitates the development of novel techniques to overcome this barrier. One of the successful strategies for combating antibiotic resistance has been the conjugation of nanoparticles (NPs) with antimicrobial moieties such as antibiotics, peptides, or other biomolecules. However, Dual Antibacterial Drug-Nanoparticle (DADNP) discovery is a slow process due to the high number of combinations of NP vs. AD compounds, assays, etc. Artificial Intelligence/Machine Learning (AI/ML) algorithms may speed up it if they predict which putative DADNP systems should be short-listed for assay. Nevertheless, the low amount of DADNP activity indicates that AI/ML analysis is tough. To solve this problem in an additive manner, the IFPTML = Information Fusion (IF) + Perturbation-Theory (PT) + Machine Learning (ML) technique was applied. Two datasets were combined (>165000 ChEMBL AD experiments with 300 NP assays) against multiple bacteria species. Eleven non-linear ML algorithms were developed using the Waikato Environment for Knowledge Analysis (WEKA) and STATISTICA. The analysis of the values of all the IFPTML models (Training/Validation Series) presents good performances (Accuracy global of 88.8-98.3%), Similarly, AUROC values are high (92-99%) in most cases. In the analysis and comparison of the algorithms used, ANN, RF, and KNN models stand out as having the highest Sn ≈ Sp ≈ 88.5% - 99.0% and AUROC ≈ 0.94 - 0.99 in both series. These results suggest that the IFPTML models may become a useful tool in the design of DADNP systems for antibacterial therapy against multidrug-resistant microbiological infections

SMILES codes are a specification in the form of online notation to describe the structure of chemical species using short ASCll (American Standard Code for Information Interchang) strings.

It contains the same information that can be found in an extended connection table, but is more useful as it is a linguistic construct rather than a computer data structure. Another important property of SMILES is that it is quite compact compared to most other methods of representing structures and involves less file space. These properties open many doors to the programmer of chemical information. For instance:

· Keys to access the database

· Mechanism for researchers to exchange chemical information

· Chemical data entry system

· Part of languages ​​for artificial intelligence or chemistry expert systems.

In this work, SMILES codes of all the compounds that participate in the intermolecular a-amidoalkylation reaction are used to calculate the molecular descriptors of the MARKOV chain, which will later be substituted in the equation from the Regression model implemented in MATEO software to predict ee(%). Therefore, in software testing the recognition and identification of SMILES is of vital importance. Furthermore, during the course of the verification and testing of the program, some weak aspects related to the programming of the software have been discovered.

In relation to the errors found in the MATEO software for the specific identification of some SMILES (Figure1), an error in the ring closure stands out for case 1. This defect occurred in the Excel learning procedure, when extending the same SMILES for the rest of the reactions. As the ending of the said SMILES is "1" Excel recognized it as a number and expanded this bug for the rest of the reactions. Although, it was a slippage caused by the experimentalist, the program was not able to particularly identify the wrong SMILES.

Although the SMILES of alkenes in this work does not pose serious problems due to their absence in the reactions studied, but they should be taken into account if this model is extended to other types of reactions.

For case 3, initially the software failed to recognize the SMILES due to the presence of the pad, which is indicative of a triple bond. This problem has been corrected and solved by Carracedo-Reboredo et al ..

Furthermore, the program does not take into account the chirality of the molecules (case 4), this has been partially remedied by multiplying the results of ee(%) by the chirality of the catalyst (+/-) 1. Although, there is no great significance in the prediction of ee(%) for the reactions studied in this work since no chiral substrates have been reported in the literature for enantioselective intermolecular a-amidoalkylation reactions. However, it is suggested to optimize the software to extend its use towards future reactions with chiral reagents.

On the other hand, different alternatives of SMILES representations were tested for the same compound, specifically the nitro group and the aromatic groups. As a result of this analysis, the software was only able to recognize the first option for the nitro group, while for the aromatic groups both alternatives were identifiable.

Finally, the possibility of SMILES recognition of non-covalently linked compounds (hydrogen bonds (case 7) and ionic bonds (case 8) was examined, since it is common to find them in the original database, either the union between the rest of solvent with the substrate, in this case it is easy to identify by the large size of the substrate compared to the solvent or the grouping between the solvent and an impurity, in this situation it is more laborious to recognize which portion refers to the solvent by the similarity of the size between these molecules. In this study, it was shown that the program was not able to consider this type of SMILES, so in a previous work a manual cleaning of them was carried out and it is also impossible to identify which part of the complex corresponds to the solvent and the reference portion to impurity. In view of this drawback, an automated cleaning by the software is proposed, since the SMILES codes can not only be used for a-amidoalkylation reactions, but it is possible to extend to other types of reactions. One way of expanding the use of MATEO consists of the development of new chemoinformatic models for the prediction of the chemical reactivity of other reactions and in this sense, although the error in the SMILES code in this master's thesis is not so important due to the scarcity of cases of non-covalently linked compounds, it is necessary to solve it as a source of SMILES code that remains for our research group.

Machine Learning (ML) is used to learn a system. One of the purposes of this automatic learning is the construction of new computational models. ML shows success in various areas such as reference systems, brain-computer interface, robotics, and chemistry.

Recently, Perturbation Theory (PT) operators and ML techniques have been combined to create powerful PTML (PT+ML) models, which are applied to complex biological systems in predicting drug-protein interaction. As for those target proteins involved in the dopamine pathway, nanotechnology, material science, etc.

The PTML models add the values ​​of the operators to the values ​​of . Therefore, we need to calculate the values ​​of the PTOs (Perturbation Theory Operators) in the data processing step. This allows us to carry out a process of merging information with variables and conditions from different sources. Moving Averages (MA), multi-condition MA (MMA), double MA, covariance operators, etc., are some examples of useful PTOs. Then, we can use Multiple Linear Regression (MLR), Linear Discriminant Analysis (LDA), or other linear ML techniques to find the PTML model. In non-linear cases, we can fit PTML models using Artificial Neural Networks (ANN), Support Vector Machines (SVM), Classification Trees, and other ML methods.

One of the important applications of ANN is found in the study of chemical reactions as an alternative to classical regression and classification techniques. Therefore, in this master's thesis, PTML-ANN models are developed with the intention of visualizing possible improvements in classical techniques. In this context, this term is presented in a generic way. ANNs are inspired by the biological neural networks of the human brain. They are made up of elements that behave in a similar way to the biological neuron in its most common functions.

The ANN aside from "resembling" the brain present a series of characteristics of the brain. For example, ANNs learn from experience, generalize from previous examples to new examples, and abstract the main characteristics of a data series. With which these networks function as interconnected neurons that create stimuli. ANN-type network is considered to be a complex biomolecular network consisting of nodes and edges. These networks are formed by "neurons", where each of them is a function, which will take a certain amount of data and provide an output response. The ANN presents three different types of functions which are:

• Input function: It is the sum function obtained by multiplying the input and output data by their weights.

• Excitation or activation function: This will take as previous input / output data. The most common types of functions depending on the input / output will be the threshold (sigmoid) or hyperbolic tangent.

• Transfer functions: It is the function used to close. The value provided by the trigger functions is the input to the ANN network.

Machine Learning (ML) is used to learn a system. One of the purposes of this machine learning is the construction of new computational models. ML shows success in various areas such as reference systems, brain-computer interface, robotics, and chemistry.

Recently, Perturbation Theory (PT) operators and ML techniques have been combined to create powerful PTML (PT+ML) models, which are applied to complex biological systems in predicting drug-protein interaction. As for those target proteins involved in the dopamine pathway, nanotechnology, material science, etc.

This PTML method has been developed by our group to search for models capable of predicting the v_ij values ​​of multiple properties of an i^th system measured under different experimental conditions c_j. In general, PTML tries to predict an objective function f(v_ij)_obs obtained from the experimental value v_ij and is obtained as a function f (v_ij) = f (s_i, c_i)_k of the structure of the system (s_i) and the conditions for a type k. PTML models can predict multiple properties of the system at the same time (multi-output and multi-objective) taking into account the variations (disturbances) with respect to a reference or expected value in multiple input variables used to quantify the experimental conditions c_j=(c₀, c₁, c₂,…. c_n) and other structural variables or molecular descriptors D_ki=(D₀, D₁, D₂,…. D_n) used to quantify the structure of the system (s_i).

The main application of this method is the study of molecular systems (drug, protein, vaccine, biomarker, nanoparticles, etc.) with multiple values ​​v_ij of parameters to optimize, which have been measured in numerous different tests with test conditions different c_j. Through this model it is possible to directly obtain the values ​​of a calculated function f(v_ij)_calc=f (s_i, c_j)_calc from a reference function f(v_ij)_ref and the disturbance operators PTO(s_i, c_j)_k.

The classification models obtain the values ​​of the probability for a specific system i in a specific test, under known test conditions c_j. The probability represents the probability of a system to be designed, it shows desired levels of the values ​​v_ij of a parameter to optimize.

For a chemical reaction, the properties to be studied could be the yield yield (%) and the enantiomeric excess ee(%). In our case we focus only on the ee(%) because our α-amidoalkylation reactions are enantioselective and their yield (%) is usually high. Furthermore, if the PTML model sought is proposed as an objective function f (v_ij) = ee(%), we would be in the presence of a regression model and the probability p(f (ee(%)_ij)=1) is not calculated. If the model attempts to classify the reactions as reactions with high excess ee(%)>cut-off or low excess ee(%) we would be in the presence of a classification model, with cut-off being a cut-off value defined by the researcher. This model would have as objective function the function f(v_ij)=f(ee(%)>cut-off))= 1 or 0. In that case if we can obtain the probability p(f(v_ij)=1)=p(f(ee(%)>cut-off)) that the system has a certain level of the ee(%)>cut-off property. In the development of the MATEO program for the regression models, the objective function f(v_ij)=ee_R(%)=dq·ee(%) was used, which quantifies the ee(%) of product R where dq= 1 when R is the majority enantiomer and dq =-1 when S is the majority enantiomer.

The PTML models add the values ​​of the operators to the values ​​of . Therefore, we need to calculate the values ​​of the PTOs (Perturbation Theory Operators) in the data processing step. This allows us to carry out a process of merging information with variables and conditions from different sources. Moving Averages (MA), multi-condition MA (MMA), double MA, covariance operators, etc., are some examples of useful PTOs. Then, we can use Multiple Linear Regression (MLR), Linear Discriminant Analysis (LDA), or other linear ML techniques to find the PTML model.

The MATEO software that we have verified is based on regression models. However, it does not implement classification models. In the case of chemical reactions, classification models are usually desirable to minimize possible errors in experimental measurements and / or to obtain a final answer on the interest of the reaction. For this reason, in addition to the regression models implemented in MATEO, we set out to develop PTML classification models. To create the PTML-LDA model, linear discriminant analysis has been used, which is a statistical classification technique that has applications to classify cases. This technique is of special interest when there are precision problems in the measurement of the observed experimental variable that make it difficult to obtain regression models, such as the ee_R(%). To use this technique, we must discretize the continuous variable ee_R(%) transforming it into a discret or Boolean variable. Initially, two alternatives were proposed for the development of the model. The first is based on the classification of the data sets into three classes, defined by the objective variable or function that takes the values ​​f(ee_R(%))_obs 1, 0, or -1. In this case, the observed function would be f(ee_R(%))_obs=1 when ee_R(%)> cut-off, otherwise f(ee_R(%))_obs = -1 if ee_R (%) <-cut- off, otherwise f(ee_R(%))_obs = 0 (cut-off> eeR (%)> -cut-off). The distribution in these groups is possible by introducing two limit values, also known as Cut-off, one of them positive and the other negative. The "-1" group is indicative of the excess S enantiomer, the "0" when it is an inefficient reaction including racemic mixtures and others (cut-off>ee_R(%)>-cut-off). The value of the observed objective function f (ee_R(%))=1 corresponds to the case of excess of R enantiomers. Furthermore, the reference function (first input variable) was not transformed in this model. Therefore, the same reference variable was used as in the previous regression models f(ee_R (%))_ref = ee_R(%)_ref.

The model obtained from this strategy is tedious and unfeasible to achieve a percentage greater than 70 for both the training and confirmation tests and achieve equitable specificity and sensitivity percentages. For this reason, this option was discarded.

The second possibility, like the previous case, makes use of the "Cut-off". This option is simpler because it allows ordering the data set into two large classes that are defined by the objective function: f(ee_R(%))_obs = 0 or 1. In the case of f(ee_R(%))_obs=0 is when the reaction has a low ee_R(%)(cut-off>ee_R(%)>-cut-off). But, when f(ee_R(%))_obs = 1 there are two sub-cases. The sub-cases are when there is an excess of R or an excess of S. In order to differentiate these two sub-groups of f(ee_R(%))_obs = 1, the reference function was modified in the input of the model. In this new PTML-LDA model the new reference function is f(ee_R(%))_ref =dq · ee_R(%)_ref.

With the STATISTICA software, it has been possible to obtain these models. This program has implemented multiple techniques for the selection of variables. The most noteworthy are "All effects", which gives the user the option to choose between the different variables that he wishes to include based on expert criteria. On the other hand, "Forward-Stepwise" makes an automatic selection of variables based on the fact that the software itself chooses the variables by doing a Fisher (F) test. For the construction of the chemoinformatic model of the present work, the first option was used, where it is possible to choose the most important perturbations in different reaction conditions. In addition, the model uses 75% of the data sets for model training and the remaining 25% for confirmation.

On the other hand, the function returned by the model belongs to the function , which provides a numerical result and coefficients of the entered variables. Additionally, to achieve an ideal model, the following points must be taken into account:

· Predicted sets should be in the 70-95% range for both the training and confirmation tests.

· The percentages of specificity (0) and sensitivity (1) must be balanced. In cases where they are not, it can lead to errors, since the model correctly predicts one of them (either the specificity or sensitivity) and the other poorly.