In hybrid gels with immobilized liquid crystal
(LC) droplets, fast and unique optical texture variations are
generated when distinct volatile organic compounds (VOCs)
interact with the LC and disturb its molecular order. The
optical texture variations can be observed under a polarized
optical microscope or transduced into a signal representing the
variations of light transmitted through the LC. We show how
hybrid gels can accurately identify 11 distinct VOCs by using
deep learning to analyze optical texture variations of individual
droplets (0.93 average F1-score) and by using machine learning
to analyze 1D optical signals from multiple droplets in hybrid
gels (0.88 average F1-score)

Electronic noses (e-noses) mimic human olfaction, by identifying Volatile Organic Compounds (VOCs). This
work presents a novel approach that successfully classifies 11 known VOCs using the signals generated by
sensing gels in an in-house developed e-nose. The proposed signals’ analysis methodology is based on the
generated signals’ morphology for each VOC since different sensing gels produce signals with different shapes
when exposed to the same VOC. For this study, two different gel formulations were considered, and an average
f1-score of 84% and 71% was obtained, respectively. Moreover, a standard method in time series classification
was used to compare the performances. Even though this comparison reveals that the morphological approach
is not as good as the 1-nearest neighbour with euclidean distance, it shows the possibility of using descriptive
sentences with text mining techniques to perform VOC classification.

Ionogel are versatile materials, as they present the electrical properties of ionic liquids and also dimensional stability, since they are trapped in a solid matrix, allowing application in electronic devices such as gas sensors and electronic noses. In this work, ionogels were designed to act as a sensitive layer for the detection of volatiles in a custom-made electronic nose. Ionogels composed of gelatin and a single imidazolium ionic liquid were doped with bare and functionalized iron oxide nanoparticles, producing ionogels with adjustable target selectivity. After exposing an array of four ionogels to 12 distinct volatile organic compounds, the collected signals were analyzed by principal component analysis (PCA) and by several supervised classification methods, in order to assess the ability of the electronic nose to distinguish different volatiles, which showed accuracy above 98%.

The materials described in this work result from the selfassembly of liquid crystals and ionic liquids into droplets,
stabilized within a biopolymeric matrix. These systems are
extremely versatile gels, in terms of composition, and offer
potential for fine tuning of both structure and function, as
each individual component can be varied. Here, the
characterization and application of these gels as sensing thin
films in gas sensor devices is presented. The unique
supramolecular structure of the gels is explored for molecular
recognition of volatile organic compounds (VOCs) by
employing gels with distinct formulations to yield
combinatorial optical and electrical responses used in the
distinction and identification of VOCs.

Affinity‐triggered assemblies rely on affinity interactions as the driving force to assemble physically‐crosslinked networks. WW domains are small hydrophobic proteins binding to proline‐rich peptides that are typically produced in the insoluble form. Previous works attempted the biological production of the full WW domain in tandem to generate multivalent components for affinity‐triggered hydrogels. In this work, an alternative approach was followed by engineering a 13‐mer minimal version of the WW domain that retains the ability to bind to target proline‐rich peptides. Both ligand and target peptides were produced chemically and conjugated to multivalent polyethylene glycol, yielding two components. Upon mixing, they together form soft biocompatible affinity‐triggered assemblies, stable in stem cell culture media, and displaying mechanical properties in the same order of magnitude as for those hydrogels formed with the full WW protein in tandem.

Artificial olfaction is a fast-growing field aiming to mimic natural olfactory systems. Olfactory systems rely on a first step of molecular recognition in which volatile organic compounds (VOCs) bind to an array of specialized olfactory proteins. This results in electrical signals transduced to the brain where pattern recognition is performed. An efficient approach in artificial olfaction combines gas-sensitive materials with dedicated signal processing and classification tools. In this work, films of gelatin hybrid gels with a single composition that change their optical properties upon binding to VOCs were studied as gas-sensing materials in a custom-built electronic nose. The effect of films thickness was studied by acquiring signals from gelatin hybrid gel films with thicknesses between 15 and 90 μm when exposed to 11 distinct VOCs. Several features were extracted from the signals obtained and then used to implement a dedicated automatic classifier based on support vector machines for data processing. As an optical signature could be associated to each VOC, the developed algorithms classified 11 distinct VOCs with high accuracy and precision (higher than 98%), in particular when using optical signals from a single film composition with 30 μm thickness. This shows an unprecedented example of soft matter in artificial olfaction, in which a single gelatin hybrid gel, and not an array of sensing materials, can provide enough information to accurately classify VOCs with small structural and functional differences.

The development of gas sensing materials is relevant in the field of non-invasive biodevices. In this work, we used an electronic nose (E-nose) developed by our research group, which possess versatile and unique sensing materials. These are gels that can be spread over the substrate by Film Coating or Spin Coating. This study aims to evaluate the influence of the sensing film spreading method selected on the classification capabilities of the E-nose. The methodology followed consisted of performing an experiment where the E-nose was exposed to 13 different pure volatile organic compounds. The sensor array had two sensing films produced by Film Coating, and other two produced by Spin Coating. After data collection, a set of features was extracted from the original signal curves, and the best were selected by Recursive Feature Elimination. Then, the classification performance of Multinomial Logistic regression, Decision Tree, and Naíve Bayes was evaluated. The results showed that both s preading methods for sensing film’s production are adequate since the estimated error of classification was inferior to 4 % for all the classification tools applied.

Electronic noses (E-noses), are usually composed by an array of sensors with different selectivities towards classes of VOCs (Volatile Organic Compounds). These devices have been applied to a variety of fields, including environmental protection, public safety, food and beverage industries, cosmetics, and clinical diagnostics. This work demonstrates that it is possible to classify eleven VOCs from different chemical classes using a single gas sensing biomaterial that changes its optical properties in the presence of VOCs. To accomplish this, an in-house built E-nose, tailor-made for the novel class of gas sensing biomaterials, was improved and combined with powerful machine learning techniques. The device comprises a delivery system, a detection system and a data acquisition and control system. It was designed to be stable, miniaturized and easy-to-handle. The data collected was pre-processed and features and curve fitting parameters were extracted from the original response. A recursive feature selection method was applied to select the best features, and then a Support Vector Machine classifier was implemented to distinguish the eleven distinct VOCs. The results show that the followed methodology allowed the classification of all the VOCs tested with 94.6% (± 0.9%) accuracy.

Electronic noses (E-noses) are devices capable of detecting and identifying Volatile Organic Compounds (VOCs) in a simple and fast method. In this work, we present the development process of an opto-electronic device based on sensing films that have unique stimuli-responsive properties, altering their optical and electrical properties, when interacting with VOCs. This interaction results in optical and electrical signals that can be collected, and further processed and analysed. Two versions of the device were designed and assembled. E-nose V1 is an optical device, and E-nose V2 is a hybrid opto-electronic device. Both E-noses architectures include a delivery system, a detection chamber, and a transduction system. After the validation of the E-nose V1 prototype, the E-nose V2 was implemented, resulting in an easy-to-handle, miniaturized and stable device. Results from E-nose V2 indicated optical signals reproducibility, and the possibility of coupling the electrical signals to the opt ical response for VOCs sensing.

Non-invasive and fast diagnostic tools based on volatolomics hold great promise in the control of infectious diseases. However, the tools to identify microbial volatile organic compounds (VOCs) discriminating between human pathogens are still missing. Artificial intelligence is increasingly recognised as an essential tool in health sciences. Machine learning algorithms based in support vector machines and features selection tools were here applied to find sets of microbial VOCs with pathogen-discrimination power. Studies reporting VOCs emitted by human microbial pathogens published between 1977 and 2016 were used as source data. A set of 18 VOCs is sufficient to predict the identity of 11 microbial pathogens with high accuracy (77%), and precision (62–100%). There is one set of VOCs associated with each of the 11 pathogens which can predict the presence of that pathogen in a sample with high accuracy and precision (86–90%). The implemented pathogen classification methodology supports future database updates to include new pathogen-VOC data, which will enrich the classifiers. The sets of VOCs identified potentiate the improvement of the selectivity of non-invasive infection diagnostics using artificial olfaction devices.

The purpose of this work was the development of a scalable and easy-to-use electronic noses (E-noses) system architecture for volatile organic compounds sensing, towards the final goal of using several E-noses acquiring large datasets at the same time. In order to accomplish this, each E-nose system is comprised by a delivery system, a detection system and a data acquisition and control system. In order to increase the scalability, the data is stored in a database common to all E-noses. Furthermore, the system was designed so it would only require five simple steps to setup a new E-nose if needed, since the only parameter that needs to be changed is the ID of the new E-nose. The user interacts with a node using an interface, allowing for the control and visualization of the experiment. At this stage, there are three different E-nose prototypes working with this architecture in a laboratory environment.

As consumption of fish and fish-based foods increases, non-destructive monitoring of fish freshness also becomes more prominent. Fish products are very perishable and prone to microbiological growth, not always easily detected by organoleptic evaluation. The analysis of the headspace of fish specimens through gas sensing is an interesting approach to monitor fish freshness. Here we report a gas sensing method for monitoring Tilapia fish spoilage based on the application of a single gas sensitive gel material coupled to an optical electronic nose. The optical signals of the sensor and the extent of bacterial growth were followed over time, and results indicated good correlation between the two determinations, which suggests the potential application of this simple and low cost system for Tilapia fish freshness monitoring.

The cooperative assembly of biopolymers and small molecules can yield functional materials with precisely tunable properties. Here, the fabrication, characterization, and use of multicomponent hybrid gels as selective gas sensors are reported. The gels are composed of liquid crystal droplets self-assembled in the presence of ionic liquids, which further coassemble with biopolymers to form stable matrices. Each individual component can be varied and acts cooperatively to tune gels' structure and function. The unique molecular environment in hybrid gels is explored for supramolecular recognition of volatile compounds. Gels with distinct compositions are used as optical and electrical gas sensors, yielding a combinatorial response conceptually mimicking olfactory biological systems, and tested to distinguish volatile organic compounds and to quantify ethanol in automotive fuel. The gel response is rapid, reversible, and reproducible. These robust, versatile, modular, pliant electro-optical soft materials possess new possibilities in sensing triggered by chemical and physical stimuli.

A common strategy for the production and purification of recombinant proteins is to fuse a tag to the protein terminal residues and employ a “tag-specific” ligand for fusion protein capture and purification. In this work, we explored the effect of two tryptophan-based tags, NWNWNW and WFWFWF, on the expression and purification of Green Fluorescence Protein (GFP) used as a model fusion protein. The titers obtained with the expression of these fusion proteins in soluble form were 0.11 mg ml−1 and 0.48 mg ml−1 for WFWFWF and NWNWNW, respectively. A combinatorial library comprising 64 ligands based on the Ugi reaction was prepared and screened for binding GFP-tagged and non-tagged proteins. Complementary ligands A2C2 and A3C1 were selected for the effective capture of NWNWNW and WFWFWF tagged proteins, respectively, in soluble forms. These affinity pairs displayed 106 M−1 affinity constants and Qmax values of 19.11 ± 2.60 ug g−1 and 79.39 ug g−1 for the systems WFWFWF AND NWNWNW, respectively. GFP fused to the WFWFWF affinity tag was also produced as inclusion bodies, and a refolding-on column strategy was explored using the ligand A4C8, selected from the combinatorial library of ligands but in presence of denaturant agents.

Adenoviruses are important platforms for vaccine development and vectors for gene therapy, increasing the demand for high titers of purified viral preparations. Monoliths are macroporous supports regarded as ideal for the purification of macromolecular complexes, including viral particles. Although common monoliths are based on synthetic polymers as methacrylates, we explored the potential of biopolymers processed by clean technologies to produce monoliths for adenovirus purification. Such an approach enables the development of disposable and biodegradable matrices for bioprocessing. A total of 20 monoliths were produced from different biopolymers (chitosan, agarose, and dextran), employing two distinct temperatures during the freezing process (−20 °C and −80 °C). The morphological and physical properties of the structures were thoroughly characterized. The monoliths presenting higher robustness and permeability rates were further analyzed for the nonspecific binding of Adenovirus serotype 5 (Ad5) preparations. The matrices presenting lower nonspecific Ad5 binding were further functionalized with quaternary amine anion-exchange ligand glycidyltrimethylammonium chloride hydrochloride by two distinct methods, and their performance toward Ad5 purification was assessed. The monolith composed of chitosan and poly(vinyl) alcohol (50:50) prepared at −80 °C allowed 100% recovery of Ad5 particles bound to the support. This is the first report of the successful purification of adenovirus using monoliths obtained from biopolymers processed by clean technologies.

A novel affinity “tag–receptor” pair was developed as a generic platform for the purification of fusion proteins. The hexapeptide RKRKRK was selected as the affinity tag and fused to green fluorescent protein (GFP). The DNA fragments were designed, cloned in Pet-21c expression vector and expressed in E. coli host as soluble protein. A solid-phase combinatorial library based on the Ugi reaction was synthesized: 64 affinity ligands displaying complementary functionalities towards the designed tag. The library was screened by affinity chromatography in a 96-well format for binding to the RKRKRK-tagged GFP protein. Lead ligand A7C1 was selected for the purification of RKRKRK fusion proteins. The affinity pair RKRKRK-tagged GFP with A7C1 emerged as a promising solution (Ka of 2.45×105 M−1). The specificity of the ligand towards the tag was observed experimentally and theoretically through automated docking and molecular dynamics simulations.

Magnetic particles {(MNPs)} offer attractive possibilities in biotechnology. {MNPs} can get close to a target biological entity, as their controllable sizes range from a few nanometres up to tens of nanometres, and their surface can be modified to add affinity and specificity towards desired molecules. Additionally, they can be manipulated by an external magnetic field gradient. In this work, the study of ferric oxide {(Fe3O4)} {MNPs} with different coating agents was conducted, particularly in terms of strategies for antibody attachment at the surfaces (covalent and physical adsorption) and the effects of blocking buffer composition and incubation times on the specific and non-specific interactions observed. The considered biological model system consisted of a coating antibody (goat {IgG)}, bovine serum albumin {(BSA)} as blocking agent, and a complementary antibody labelled with {FITC} (anti-goat {IgG).} The detection of antibody binding was followed by fluorescence microscopy and the intensity of the signals quantified. The ratio between the mean grey values of negative and positive controls, as well as the maximum intensity attainable in positive controls, were considered in the evaluation of the assays efficiency. The covalent immobilization of the coating antibody was more successful as opposed to protein adsorption. For covalent immobilization, silica-coated {MNPs}, a 5% (w/v) concentration of {BSA} in the blocking buffer and incubation times of 1 h produced the best results in terms of assay sensitivity. However, when conducting the assay for incubation periods of 10 min, the fluorescence signal was reduced by 44% but the assay specificity was maintained.