Theory and rationale of interpretable all-in-one pattern discovery and disentanglement system.

In machine learning (ML), association patterns in the data, paths in decision trees, and weights between layers of the neural network are often entangled due to multiple underlying causes, thus masking the pattern-to-source relation, weakening prediction, and defying explanation. This paper presents a revolutionary ML paradigm: pattern discovery and disentanglement (PDD) that disentangles associations and provides an all-in-one knowledge system capable of (a) disentangling patterns to associate with distinct primary sources; (b) discovering rare/imbalanced groups, detecting anomalies and rectifying discrepancies to improve class association, pattern and entity clustering; and (c) organizing knowledge for statistically supported interpretability for causal exploration. Results from case studies have validated such capabilities. The explainable knowledge reveals pattern-source relations on entities, and underlying factors for causal inference, and clinical study and practice; thus, addressing the major concern of interpretability, trust, and reliability when applying ML to healthcare, which is a step towards closing the AI chasm.

Effects of LED exposure on the sensory properties of soybean oil.

Soybean oil is an important commodity in the US and abroad. As lipids are vulnerable to oxidation, damage to soybean oil has been previously reported to arise from prolonged light exposure. Here, we assessed the extent of damage to samples of soybean oil from light, tested with sensory evaluation, assaying the effects of both LED and fluorescent lighting (both at 2000 lux) exposure to oil in PET bottles, using a variety of sensory approaches. Threshold testing (n = 66) suggested that differences in the sensory properties of soybean oil could be detected by a human panelist after only 32 h of LED exposure. Sensory flash profiling (n = 21) again confirmed that light-exposed samples were perceived differently to shielded samples, with LED-exposed samples clustering together, but separately from those instead exposed to fluorescent light. These differences did not necessarily result in a drop in liking of the samples, assessed with consumer testing (n = 94), despite a trend for lower liking in exposed versus shielded samples. With similar levels of light exposure, elevated dissolved oxygen levels could be mitigated with O2 scavenging. Our results suggest that soybean oil may be reaching the consumer at a differing quality than intended, which may be mitigated with superior packaging technologies. PRACTICAL APPLICATION: Soybean oil is used widely in the US and global food systems; however, it can be damaged via light oxidation. Here, we show that human panelists can readily detect the sensory signals from light damage in oil exposed for only short periods of time, via either fluorescent or LED light sources. The results suggest that superior soybean oil could be delivered to the market if improvements in packaging technologies were considered.

Immunomodulatory activity of Senegalia macrostachya (Reichenb. ex DC.) Kyal. & Boatwr seed polysaccharide fraction through the activation of the MAPK signaling pathway in RAW264.7 macrophages.

Senegalia macrostachya (Reichenb. ex DC.) Kyal. & Boatwr seed (SMS) is a wild legume used as food and medicine in many African countries. In the current study, a novel polysaccharide (SMSP2) was extracted from SMS using hot water and purified with DEAE-52 cellulose. Its structure was characterized, and the immunomodulatory activity and possible molecular mechanism in murine macrophage RAW264.7 were explored. The results revealed that SMSP2 was a uronic acid-rich polysaccharide (51.6%, w/w) with a molecular weight of 52.07 kDa. The neutral sugars were mainly arabinose, xylose, mannose, and galactose at a molar ratio of 1.00 : 0.84 : 0.90 : 0.07. Interestingly, SMSP2 treatment markedly promoted macrophage proliferation and phagocytosis and induced the expression of inflammatory mediators, such as nitric oxide (NO), tumor necrosis factor-α (TNF-α), interleukin (IL)-1ß, IL-6, and IL-10. SMSP2-induced macrophage stimulation occurs through the activation of the mitogen-activated protein kinase (MAPK) signaling pathway. Moreover, macrophage surface complement receptor 3 (CR3) might play an important role in SMSP2-induced macrophage activation. This study revealed that SMSP2 is a potent immunomodulator, which could be used as a functional food and a pharmaceutical adjuvant in treating immune-compromising diseases.

Pattern discovery and disentanglement on relational datasets.

Machine Learning has made impressive advances in many applications akin to human cognition for discernment. However, success has been limited in the areas of relational datasets, particularly for data with low volume, imbalanced groups, and mislabeled cases, with outputs that typically lack transparency and interpretability. The difficulties arise from the subtle overlapping and entanglement of functional and statistical relations at the source level. Hence, we have developed Pattern Discovery and Disentanglement System (PDD), which is able to discover explicit patterns from the data with various sizes, imbalanced groups, and screen out anomalies. We present herein four case studies on biomedical datasets to substantiate the efficacy of PDD. It improves prediction accuracy and facilitates transparent interpretation of discovered knowledge in an explicit representation framework PDD Knowledge Base that links the sources, the patterns, and individual patients. Hence, PDD promises broad and ground-breaking applications in genomic and biomedical machine learning.

Explanation and prediction of clinical data with imbalanced class distribution based on pattern discovery and disentanglement.

BACKGROUND: Statistical data analysis, especially the advanced machine learning (ML) methods, have attracted considerable interest in clinical practices. We are looking for interpretability of the diagnostic/prognostic results that will bring confidence to doctors, patients and their relatives in therapeutics and clinical practice. When datasets are imbalanced in diagnostic categories, we notice that the ordinary ML methods might produce results overwhelmed by the majority classes diminishing prediction accuracy. Hence, it needs methods that could produce explicit transparent and interpretable results in decision-making, without sacrificing accuracy, even for data with imbalanced groups. METHODS: In order to interpret the clinical patterns and conduct diagnostic prediction of patients with high accuracy, we develop a novel method, Pattern Discovery and Disentanglement for Clinical Data Analysis (cPDD), which is able to discover patterns (correlated traits/indicants) and use them to classify clinical data even if the class distribution is imbalanced. In the most general setting, a relational dataset is a large table such that each column represents an attribute (trait/indicant), and each row contains a set of attribute values (AVs) of an entity (patient). Compared to the existing pattern discovery approaches, cPDD can discover a small succinct set of statistically significant high-order patterns from clinical data for interpreting and predicting the disease class of the patients even with groups small and rare. RESULTS: Experiments on synthetic and thoracic clinical dataset showed that cPDD can 1) discover a smaller set of succinct significant patterns compared to other existing pattern discovery methods; 2) allow the users to interpret succinct sets of patterns coming from uncorrelated sources, even the groups are rare/small; and 3) obtain better performance in prediction compared to other interpretable classification approaches. CONCLUSIONS: In conclusion, cPDD discovers fewer patterns with greater comprehensive coverage to improve the interpretability of patterns discovered. Experimental results on synthetic data validated that cPDD discovers all patterns implanted in the data, displays them precisely and succinctly with statistical support for interpretation and prediction, a capability which the traditional ML methods lack. The success of cPDD as a novel interpretable method in solving the imbalanced class problem shows its great potential to clinical data analysis for years to come.

In vitro gastric emptying characteristics of konjac glucomannan with different viscosity and its effects on appetite regulation.

Konjac glucomannan (KGM) is associated with the satiety-enhancing property by imparting the food matrix with high viscosity. In the present study, rheology tests on KGM sol with different viscosities were conducted to understand its flow behavior as they presented in the mouth and stomach, and the in vitro gastric emptying characteristics of KGM were examined with a human gastric simulator. Then, their effects on subjective appetite, glycemia, and appetite-related hormones (insulin, GLP-1, PYY3-36, CCK-8, ghrelin) response were investigated by conducting a randomized, single-blind, crossover trial in 22 healthy adults (11 female and 11 male, mean age (years): 23.2 ± 2.0, BMI (kg m-2): 20.6 ± 2.1). The blood samples and ratings for subjective appetite were collected at regular time intervals after the subjects were fed with four test breakfasts (one control treatment and three experimental treatments) on four different days. An ad libitum lunch was provided to the subjects once they consumed the breakfasts and their food intake was recorded. As the viscosity increased, the gastric emptying rate was delayed despite a large part of the chyme viscosity lost during digestion. The satiating capacity of the test breakfast was significantly enhanced as its viscosity increased, the and subjects' sensation for hunger, fullness, desire-to-eat, and prospective food consumption differed significantly (p = 0.006, 0.000, 0.002, and 0.001, respectively) between the treatments. The secretion of glycemia and satiety-related hormones were beneficially modulated by the increased viscosity of the test meal but a small decrease in the ad libitum food intake was observed after the intervention of the viscous test breakfasts. Overall, elevating the meal viscosity moderately by using KGM could contribute to combating the challenge of hunger for people in the bodyweight management.

Carboxymethylpachymaran-zein coated plant microcapsules-based ß-galactosidase encapsulation system for long-term effective delivery.

Oral delivery of ß-galactosidase (ß-Gal) for alleviating lactose intolerance is a challenge due to its degradation in the harsh gastrointestinal tract (GIT). In this study, sunflower sporopollenin exine capsules (SECs)-based systems were fabricated to serve as protective microcapsules in order to improve the stability as well as to impact the pH-responsive release of ß-Gal in GIT. The SECs extraction and loading process were optimized with the maximum residual activity of 82.75 ± 2.16%. Furthermore, the comparison of two systems indicated that ß-Gal loaded into SECs which were entrapped in CMP-zein system was superior to that in zein system in terms of delivering ß-Gal to the intestinal tract. Additionally, the interaction between CMP and zein confirmed by FTIR might contribute to the increased resistance to the GIT. Collectively, these results suggested that SECs-based CMP-zein system might be useful for encapsulation, protection, and delivery of bioactive substances.

Development of multi-layered gastric floating tablets based on konjac glucomannan: a modified calcium supplement with enhanced bioavailability.

This study developed a gastric-floating delivery system of calcium based on konjac glucomannan (KGM). The developed calcium tablets, consisting of one core layer coated with two barrier layers, were fabricated with a facile method. The role of KGM within the tablets was evaluated from characteristics including swelling behavior, hydrated gel properties, floating ability and release profiles of calcium (Ca). The results indicated that upregulating the KGM fraction accelerated the formation of a more compact gel network in gastric conditions, which prolonged both the floating lag time and floating duration, which resulted in a more sustained swelling behavior and a slower release of calcium. Among all the formulations, a core tablet containing 20% KGM (K20) was selected as the optimized one as it could quickly float up in 7.21 s, exhibited an almost linear release and obtained a release amount of 87.73% within 12 h. Finally, a comparison of in vivo calcium bioavailability between a KGM-based calcium tablet and a commercial calcium tablet (Caltrate®) was carried out by monitoring the serum calcium concentration after administration in rabbits. The results suggested that, after having the KGM-based calcium tablet, the changes of serum calcium levels were gentler due to a sustained-release property. The difference integral value between profile K20 and the baseline was 1.4358, larger than that of Caltrate® (1.1808), suggesting the higher absorption efficiency of KGM-based calcium tablets.

Discovery and disentanglement of aligned residue associations from aligned pattern clusters to reveal subgroup characteristics.

BACKGROUND: A protein family has similar and diverse functions locally conserved. An aligned pattern cluster (APC) can reflect the conserved functionality. Discovering aligned residue associations (ARAs) in APCs can reveal subtle inner working characteristics of conserved regions of protein families. However, ARAs corresponding to different functionalities/subgroups/classes could be entangled because of subtle multiple entwined factors. METHODS: To discover and disentangle patterns from mixed-mode datasets, such as APCs when the residues are replaced by their fundamental biochemical properties list, this paper presents a novel method, Extended Aligned Residual Association Discovery and Disentanglement (E-ARADD). E-ARADD discretizes the numerical dataset to transform the mixed-mode dataset into an event-value dataset, constructs an ARA Frequency Matrix and then converts it into an adjusted Statistical Residual (SR) Vector Space (SRV) capturing statistical deviation from randomness. By applying Principal Component (PC) Decomposition on SRV, PCs ranked by their variance are obtained. Finally, the disentangled ARAs are discovered when the projections on a PC is re-projected to a vector space with the same basis vectors of SRV. RESULTS: Experiments on synthetic, cytochrome c and class A scavenger data have shown that E-ARADD can a) disentangle the entwined ARAs in APCs (with residues or biochemical properties), b) reveal subtle AR clusters relating to classes, subtle subgroups or specific functionalities. CONCLUSIONS: E-ARADD can discover and disentangle ARs and ARAs entangled in functionality and location of protein families to reveal functional subgroups and subgroup characteristics of biological conserved regions. Experimental results on synthetic data provides the proof-of-concept validation on the successful disentanglement that reveals class-associated ARAs with or without class labels as input. Experiments on cytochrome c data proved the efficacy of E-ARADD in handing both types of residue data. Our novel methodology is not only able to discover and disentangle ARs and ARAs in specific statistical/functional (PCs and RSRVs) spaces, but also their locations in the protein family functional domains. The success of E-ARADD shows its great potential to proteomic research, drug discovery and precision and personalized genetic medicine.

Effects of thermal sterilization on soy protein isolate/polyphenol complexes: Aspects of structure, in vitro digestibility and antioxidant activity.

The main purpose of this work was to investigate the influence of typical thermal sterilization approaches (pasteurization, high-temperature sterilization) on the structure, in vitro digestibility, and antioxidant activity of soy protein isolate (SPI)/black soybean seed coat extract (BE) complexes at pH 7.0. The results of zeta potential and particle size demonstrated that the addition of BE was contributed to inhibit protein thermal aggregation. Heat sterilization resulted in protein unfolding revealed by UV-Vis, circular dichroism, and fluorescence spectroscopy analysis. The increase of BE led to the reduction of fluorescence intensity and surface hydrophobicity of SPI. Moreover, the formation of SPI/BE complexes inhibited protein digestion in vitro, while thermal sterilization promoted protein digestion. The SPI/BE complexes showed a strong radical scavenging ability both before and after thermal treatment.

Revealing Subtle Functional Subgroups in Class A Scavenger Receptors by Pattern Discovery and Disentanglement of Aligned Pattern Clusters.

A protein family has similar and diverse functions locally conserved as aligned sequence segments. Further discovering their association patterns could reveal subtle family subgroup characteristics. Since aligned residues associations (ARAs) in Aligned Pattern Clusters (APCs) are complex and intertwined due to entangled function, factors, and variance in the source environment, we have recently developed a novel method: Aligned Residue Association Discovery and Disentanglement (ARADD) to solve this problem. ARADD first obtains from an APC an ARA Frequency Matrix and converts it to an adjusted statistical residual vectorspace (SRV). It then disentangles the SRV into Principal Components (PCs) and Re-projects their vectors to a SRV to reveal succinct orthogonal AR groups. In this study, we applied ARADD to class A scavenger receptors (SR-A), a subclass of a diverse protein family binding to modified lipoproteins with diverse biological functionalities not explicitly known. Our experimental results demonstrated that ARADD can unveil subtle subgroups in sequence segments with diverse functionality and highly variable sequence lengths. We also demonstrated that the ARAs captured in a Position Weight Matrix or an APC were entangled in biological function and domain location but disentangled by ARADD to reveal different subclasses without knowing their actual occurrence positions.

Response of vitamins A, E, hematological and serum biochemical markers in Crucian carp (Carassius auratus gibelio) exposed to environmental Pb(2+) and Cd(2+).

Toxicity of Pb(2+) and Cd(2+) is a widespread issue in the world; however, few studies have been conducted to understand their effect at environmentally realistic concentration in a mixture. In the present study, Crucian carp was exposed to Pb(2+) (30 µg·l(-1)), Cd(2+) (100 µg·l(-1)) and their mixture (30+100 µg·l(-1)) for 96 h and 21 d period to assess changes in the liver and muscle vitamin A and E content, and hematological and serum biochemical parameters. The results indicated significant decline in the level of antioxidant vitamins A, E and alterations in the hematological and serum biochemical indices. The toxicity revealed anemia, impairment of the liver and kidney with evident responses after 21 d exposure due to additive effect of Pb(2+) and Cd(2+) in mixture. Moreover, the differential response of vitamins A, E and blood parameters to low levels of waterborne Pb(2+) and Cd(2+) in freshwater fish can be used as biomarkers for monitoring contamination of aquatic environment.

[Study on the epidemiology and determinants of drug-resistant tuberculosis in northern rural area of Jiangsu province].

OBJECTIVE: To understand the determinants and epidemiology of drug-resistant tuberculosis (TB) in rural area. METHODS: All the diagnosed TB patients in a county with directly observed treatment (DOTS) short-course program in 2002 and a sample of patients in another county without DOTS program located in northern Jiangsu province were surveyed with questionnaires. Drug susceptibility testing (DST) for positive cultures were performed by standardized proportion method. Univariable analysis and multivariate nonconditional logistic regression modeling were applied for data analysis. RESULTS: Among the 152 patients with DST results, 32.9% of the cases showed resistance to at least one of the first-line anti-tuberculosis drugs with 26.3% to isoniazid, 18.4% to rifampin and 17.1% to both isoniazid and rifampin respectively. Previous treatments for TB and residence in the county without DOTS program were independent risk factors for isoniazid and rifampin resistance. TB patients showing indifferent to their health and delayed health seeking for more than 1 month were more likely to have rifampin resistance. Independent predictors of multidrug-resistant TB would include delayed health seeking for more than 1 month (OR = 4.66, 95% CI: 1.26 - 17.24), residing in the county without a DOTS program (OR = 3.01, 95% CI: 1.10 - 8.22), indifference to their health condition (OR = 5.13, 95% CI: 1.06 - 24.90) and suffering from chronic diseases (OR = 0.22, 95% CI: 0.05 - 0.87). CONCLUSION: Drug-resistant TB was quite serious in this rural areas, mainly associated with man-made factors but partly due to the availability of the transmission.