In the field of oculography tracking reading progress is challenging due to measurement errors in eye tracking systems. This paper presents a two-stage approach using an autoencoding neural network model. The model uses fixation duration and text structure to create a probability map representing the likelihood that each pixel is viewed and interprets this map using a specific criterion to determine whether a word has been read. The model was trained and evaluated using a synthetic dataset generated from the ZuCo 1.0 dataset. The model achieved an F-measure value of 0.9782 and an MIoU value of 0.9587 on a test set of 537 pages for the task of classifying words into read and unread.

Tyrosinase amperometric biosensors based on graphite printed electrodes modified with reduced graphene oxide (RGO), multi-walled carbon nanotubes (MWNTs), and a nanocomposite based on RGO and silver nanoparticles (Ag NPs), as well as a variant of polarization fluorescence immunoanalysis (PFIA) using a tracer based on fluorescein-5(6)-carboxamidohexanoic acid for determining aristolochic acid Ⅰ (AA) were developed. It has been found that AA Ⅰ is a tyrosinase inhibitor in the concentration range of 1×10–10–1×10–8 M, with LOD of 7×10–11 M. Modifying the electrode surface by RGO and MWCNTs improved the analytical characteristics of the biosensor; analyzed concentration range increased to 1×10–11–1×10–6 M for RGO and 1×10–10–1×10–6 M for MWCNTs. The correlation coefficient and LOD were 0.9828 and 8×10–12 M, and 0.9859 and 5×10–11 M in the case of the biosensors modified with RGO and MWCNT, respectively. A variant of a competitive polarization fluorescent immunoanalysis was developed, which permitted determining AA Ⅰ in the concentration range of 1×10–11–1×10–7 M with LOD of 7×10–12 M. The concentration of antibodies was 1 mg/ml, the time of incubation of the immune complex tracer-antibody was 5 min. Methods for determining AA Ⅰ were tested using samples of herbal preparations, in roots, leaves, stems of European hoof, as well as in crops grown together with hoof.

To ensure express control of food quality and safety, we developed an immunochromatographic test system for detection of the antibiotic clinafloxacin. The test system implements an indirect competitive immunoassay format based on the interaction of unmodified specific polyclonal antibodies with the antibiotic potentially present in the sample and the antibiotic–protein conjugate immobilized on the working membrane of the test strip. The detection of immune complexes formed on the membrane is carried out using a conjugate of anti-species antibodies and gold nanoparticles. We determined immunochromatography conditions that ensure the achievement of the minimum detection limit. At the assay duration of 20 min, the instrumental and visual detection limits of clinafloxacin comprise 0.3 and 10 ng/mL, respectively. The selectivity of the test system in relation to antibiotics from the class of fluoroquinolones was studied. To ensure the control of the contamination of honey with clinafloxacin and structurally related fluoroquinolones, we identified an optimal sample preparation regimen and confirmed the efficacy of the developed test system.

A method for preparing antibodies that specifically recognize antibiotics of the fluoroquinolone group with the same radical in the first position of the quinolone nucleus is proposed. The specificity of rabbit antisera prepared at different cycles of immunization by changing the structure of the hapten in the composition of immunogens was characterized. The selected antibodies provided a group-specific analysis of 16 representatives of fluoroquinolones, including a combination of the following compounds that are controlled in animal products: ciprofloxacin, norfloxacin, pefloxacin, ofloxacin, and enrofloxacin. Using these antibodies, an indirect competitive enzyme-linked immunosorbent assay was developed that was characterized by a detection limit of ciprofloxacin of 0.2 ng/mL and a duration of 2 h. The assay was approbated for the detection of fluoroquinolones in milk.

BACKGROUND Ciprofloxacin (CIP) and chloramphenicol (CAP) are relevant antibiotics of the fluoroquinolone (FQ) and amphenicol (AP) groups, respectively, widely used in veterinary practice and they contaminate agricultural products. In this study, a rapid and sensitive immunochromatographic assay (ICA) was developed for simultaneous detection of CIP and CAP in dairy products. The ICA was carried out in a direct competitive format using gold nanoparticles as a label. RESULTS The ICA developed here allowed for the detection of CIP and CAP in Triton X-100-containing buffered saline (PBST) within 15 min with instrumental detection limits of 20 pg mL(-1) and 0.5 ng mL(-1), respectively, and with a visual detection limit of 5 ng mL(-1) for both antibiotics. The ICA showed cross-reactivity (69-160%) to 19 antibiotics in the FQ group and no cross-reactivity (<0.1%) to 2 antibiotics of the AP group. The ICA allowed detection of CIP and CAP in a panel of dairy products by employing a simple procedure of preliminary sample preparation. The detection limits for the two antibiotics were the same as in PBST. The analytical recoveries of CIP and CAP in dairy products ranged from 83% to 120%. CONCLUSION The analytical characteristics of the test system allow its use for the detection of antibiotics in milk and dairy products during all steps of production. (c) 2019 Society of Chemical Industry

A common problem in the immunodetection of structurally close compounds is understanding the regularities of immune recognition, and elucidating the basic structural elements that provide it. Correct identification of these elements would allow for select immunogens to obtain antibodies with either wide specificity to different representatives of a given chemical class (for class-specific immunoassays), or narrow specificity to a unique compound (mono-specific immunoassays). Fluoroquinolones (FQs; antibiotic contaminants of animal-derived foods) are of particular interest for such research. We studied the structural basis of immune recognition of FQs by antibodies against ciprofloxacin (CIP) and clinafloxacin (CLI) as the immunizing hapten. CIP and CLI possess the same cyclopropyl substituents at the N1 position, while their substituents at C7 and C8 are different. Anti-CIP antibodies were specific to 22 of 24 FQs, while anti-CLI antibodies were specific to 11 of 26 FQs. The molecular size was critical for the binding between the FQs and the anti-CIP antibody. The presence of the cyclopropyl ring at the N1 position was important for the recognition between fluoroquinolones and the anti-CLI antibody. The anti-CIP quantitative structure–activity relationship (QSAR) model was well-equipped to predict the test set (pred_R2 = 0.944). The statistical parameters of the anti-CLI model were also high (R2 = 0.885, q2 = 0.864). Thus, the obtained QSAR models yielded sufficient correlation coefficients, internal stability, and predictive ability. This work broadens our knowledge of the molecular mechanisms of FQs’ interaction with antibodies, and it will contribute to the further development of antibiotic immunoassays.

In this study, highly sensitive immunochromatographic analyses (ICAs) of ciprofloxacin (CIP) are developed. CIP is an antibiotic of the fluoroquinolone group that is widely used in veterinary practice and contaminates agricultural products. The analyses are based on various techniques that introduce gold nanoparticles as markers. It has been shown that the schemes allow for the detection of CIP within 15 min with an instrumental detection limit of 10 pg/mL for both schemes and visual detection limits of 10 and 2ng/mL for direct and indirect schemes, respectively. The developed systems have been tested to detect the antibiotic in milk samples. It has been shown that ICA can determine CIP in milk using a simple procedure of preliminary sample preparation with preservation of high analytical characteristics.

Изучены условия получения распознающего слоя пьезоэлектрического сенсора на основе многостенных углеродных нанотрубок (УНТ) для высокочувствительного определения фторхинолонов в прямом и конкурентном форматах иммуноанализа. Установлено, что использование на стадии иммобилизации УНТ способствует увеличению присоединенной массы и концентрационной чувствительности сенсора, вследствие возникновения 3D граничного слоя, повышающего связывающую способность поверхности электрода, и, следовательно, эффективности аффинных взаимодействий для анализируемых фторхинолонов. Показано, что применение УНТ при формировании распознающего слоя расширяет диапазон определяемых содержаний фторхинолонов в конкурентном (10-350 и 10-370 нг/см3 для левофлоксацина и ципрофлоксацина) и прямом форматах анализа (30-650 и 25-670 нг/см3 для левофлоксацина и ципрофлоксацина).

Two selective immunotechniques based on heterologous approach were proposed and chacterized to detect danofloxacin, a priority veterinary contaminant, in milk. The first technique is fluorescence polarisation immunoassay (PFIA), where a danofloxacin tracer was synthesised with 4-aminomethyl fluorescein, with polyclonal antibodies obtained for a danofloxacin conjugate with cationised bovine serum albumin. The developed system has a detection limit for danofloxacin equal to 13 ng/mL and does not show any cross-reactivity with other 24 tested fluoroquinolones. The second assay format is enzymelinked immunosorbent assay (ELISA) for danofloxacin that is based on the application of the same antibodies and a heterologous solid-phase clinafloxacin conjugate with ovalbumin. The limit of danofloxacin detection for the given ELISA is 0.5 ng/mL, and its strong selectivity to danofloxacin has been also confirmed. The possibility of quantitative control of danofloxacin content in milk has been shown for the developed ELISA technique; the opening percentage was in the range from 83 to 130%.

BACKGROUND Clinafloxacin is used for the treatment of disease in food-producing animals, e.g. Brucella melitensis, which often occurs in goats; however, the clinafloxacin residue in goat milk may harm human health and result in the development of drug-resistant bacterial strains or allergies. Despite this, there is not a rapid, sensitive and accurate analytical method in goat milk for rapid screening or monitoring purposes. RESULTS One homologous and five heterologous tracers were designed and compared for fluorescence polarization immunoassay (FPIA) optimization. Based on the combination of a heterologous tracer (PAZ-FITC, synthesized with pazufloxacin and FITC) and the antibody against clinafloxacin, a highly sensitive FPIA was established for the detection of clinafloxacin residue in goat milk for the first time. The IC50 value was 29.3 µg L(-1) for clinafloxacin in the heterologous format - six times lower than that of the combination of the homologous tracers and the antibody. The recoveries ranged from 86.8% to 104.5%, with the relative standard deviation ranging from 4.1% to 7.2%. Validation by high-performance liquid chromatography (HPLC) confirmed that the results obtained from the proposed FPIA were in agreement with those of HPLC. CONCLUSION This proposed heterologous strategy for enhanced FPIA is sensitive and rapid enough for the high-throughput detection of clinafloxacin residue in goat milk.

A method of fluorescence polarization immunoassay (FPIA) has been developed for the determination of levofloxacin in urine. Fluorescein-tagged antigens (tracers) with various fluoroquinolones and fluorescent tags have been synthesized. The tracer based on garenoxacin labeled with 4-aminomethylfluorescein and polyclonal antibodies to levofloxacin was the most sensitive for the determination of fluoroquinolones. A FPIA method has been developed for the determination of levofloxacin with the limit of detection 1.0 ng/mL and analytical range from 2.5 to 50 ng/mL. A one-reagent FPIA method has been developed for the determination of levofloxacin with the limit of detection 0.5 ng/mL and analytical range from 1 to 10 ng/mL. The methods have been tested in the determination of levofloxacin in urine.

In this work we obtained polyclonal antibodies for levofloxacin. We synthesized conjugates of levofloxacin with cationized BSA for immunization and with ovalbumin for development of an ELISA for the detection of levofloxacin. The method is characterized by a range of detectable concentrations of 0.03 ng/mL to 0.41 ng /mL and the limit of detectable concentrations is 0.01 ng/mL. We tested the 28 fluoroquinolones for cross reactivity and only ofloxacin (145%), marbofloxacin (82%), ofloxacin in its dextrorotatory form (68%), rufloxacin (67%), and garenoxacin (24%) had cross reactivity. The optimized ELISA technique allowed the detection of levofloxacin in milk from 0.33 ng/mL to 3.34 ng/mL. The recoveries were in the range of 89.5–102% with a relative standard deviation of 3%. We tested 45 real samples of milk purchased in local stores; for 5 of them the results were positive (near 1 ng/mL).