Enhancing post-traumatic stress disorder patient assessment: leveraging natural language processing for research of domain criteria identification using electronic medical records.

BACKGROUND: Extracting research of domain criteria (RDoC) from high-risk populations like those with post-traumatic stress disorder (PTSD) is crucial for positive mental health improvements and policy enhancements. The intricacies of collecting, integrating, and effectively leveraging clinical notes for this purpose introduce complexities. METHODS: In our study, we created a natural language processing (NLP) workflow to analyze electronic medical record (EMR) data and identify and extract research of domain criteria using a pre-trained transformer-based natural language model, all-mpnet-base-v2. We subsequently built dictionaries from 100,000 clinical notes and analyzed 5.67 million clinical notes from 38,807 PTSD patients from the University of Pittsburgh Medical Center. Subsequently, we showcased the significance of our approach by extracting and visualizing RDoC information in two use cases: (i) across multiple patient populations and (ii) throughout various disease trajectories. RESULTS: The sentence transformer model demonstrated high F1 macro scores across all RDoC domains, achieving the highest performance with a cosine similarity threshold value of 0.3. This ensured an F1 score of at least 80% across all RDoC domains. The study revealed consistent reductions in all six RDoC domains among PTSD patients after psychotherapy. We found that 60.6% of PTSD women have at least one abnormal instance of the six RDoC domains as compared to PTSD men (51.3%), with 45.1% of PTSD women with higher levels of sensorimotor disturbances compared to men (41.3%). We also found that 57.3% of PTSD patients have at least one abnormal instance of the six RDoC domains based on our records. Also, veterans had the higher abnormalities of negative and positive valence systems (60% and 51.9% of veterans respectively) compared to non-veterans (59.1% and 49.2% respectively). The domains following first diagnoses of PTSD were associated with heightened cue reactivity to trauma, suicide, alcohol, and substance consumption. CONCLUSIONS: The findings provide initial insights into RDoC functioning in different populations and disease trajectories. Natural language processing proves valuable for capturing real-time, context dependent RDoC instances from extensive clinical notes.

Epicatechin-Promoted Formation of Acrylamide from 3-Aminopropionamide Via Postoxidative Reaction of B-Ring.

The role of thermally generated 3-aminopropionamide as an intermediate in acrylamide formation in the Maillard reaction has been well established. Herein, the effect of epicatechin on the conversion of 3-aminopropionamide into acrylamide under oxidative conditions was investigated at 160-220 °C. Epicatechin promoted acrylamide generation and 3-aminopropionamide degradation. The stable isotope-labeling technique combined with UHPLC-Orbitrap-MS/MS analysis showed adduct formation between 3-aminopropionamide and the oxidized B ring of epicatechin to form a Schiff base. This initially formed Schiff base could directly degrade to acrylamide, undergo reduction or dehydration to other intermediates, and subsequently generate acrylamide. Based on accurate mass analysis, five intermediates with intact or dehydrated C rings were tentatively identified. Furthermore, reaction pathways were proposed that were supported by the changes in the levels of adducts formed during heating. To the authors' knowledge, this study is the first to reveal pathways through which flavanols promoted the formation of acrylamide in Maillard reactions.

Enhancing post-traumatic stress disorder patient assessment: Leveraging Natural Language Processing for Research of Domain Criteria Identification using electronic medical records.

Background: Extracting research of domain criteria (RDoC) from high-risk populations like those with post-traumatic stress disorder (PTSD) is crucial for positive mental health improvements and policy enhancements. The intricacies of collecting, integrating, and effectively leveraging clinical notes for this purpose introduce complexities. Methods: In our study, we created an NLP workflow to analyze electronic medical record (EMR) data, and identify and extract research of domain criteria using a pre-trained transformer-based natural language model, allmpnet-base-v2. We subsequently built dictionaries from 100,000 clinical notes and analyzed 5.67 million clinical notes from 38,807 PTSD patients from the University of Pittsburgh Medical Center. Subsequently, we showcased the significance of our approach by extracting and visualizing RDoC information in two use cases: (i) across multiple patient populations and (ii) throughout various disease trajectories. Results: The sentence transformer model demonstrated superior F1 macro scores across all RDoC domains, achieving the highest performance with a cosine similarity threshold value of 0.3. This ensured an F1 score of at least 80% across all RDoC domains. The study revealed consistent reductions in all six RDoC domains among PTSD patients after psychotherapy. Women had the highest abnormalities of sensorimotor systems, while veterans had the highest abnormalities of negative and positive valence systems. The domains following first diagnoses of PTSD were associated with heightened cue reactivity to trauma, suicide, alcohol, and substance consumption. Conclusions: The findings provide initial insights into RDoC functioning in different populations and disease trajectories. Natural language processing proves valuable for capturing real-time, context dependent RDoC instances from extensive clinical notes.

Prediction of adverse events risk in patients with comorbid post-traumatic stress disorder and alcohol use disorder using electronic medical records by deep learning models.

BACKGROUND: Identifying co-occurring mental disorders and elevated risk is vital for optimization of healthcare processes. In this study, we will use DeepBiomarker2, an updated version of our deep learning model to predict the adverse events among patients with comorbid post-traumatic stress disorder (PTSD) and alcohol use disorder (AUD), a high-risk population. METHODS: We analyzed electronic medical records of 5565 patients from University of Pittsburgh Medical Center to predict adverse events (opioid use disorder, suicide related events, depression, and death) within 3 months at any encounter after the diagnosis of PTSD+AUD by using DeepBiomarker2. We integrated multimodal information including: lab tests, medications, co-morbidities, individual and neighborhood level social determinants of health (SDoH), psychotherapy and veteran data. RESULTS: DeepBiomarker2 achieved an area under the receiver operator curve (AUROC) of 0.94 on the prediction of adverse events among those PTSD+AUD patients. Medications such as vilazodone, dronabinol, tenofovir, suvorexant, modafinil, and lamivudine showed potential for risk reduction. SDoH parameters such as cognitive behavioral therapy and trauma focused psychotherapy lowered risk while active veteran status, income segregation, limited access to parks and greenery, low Gini index, limited English-speaking capacity, and younger patients increased risk. CONCLUSIONS: Our improved version of DeepBiomarker2 demonstrated its capability of predicting multiple adverse event risk with high accuracy and identifying potential risk and beneficial factors.

DeepBiomarker2: Prediction of Alcohol and Substance Use Disorder Risk in Post-Traumatic Stress Disorder Patients Using Electronic Medical Records and Multiple Social Determinants of Health.

Prediction of high-risk events amongst patients with mental disorders is critical for personalized interventions. We developed DeepBiomarker2 by leveraging deep learning and natural language processing to analyze lab tests, medication use, diagnosis, social determinants of health (SDoH) parameters, and psychotherapy for outcome prediction. To increase the model's interpretability, we further refined our contribution analysis to identify key features by scaling with a factor from a reference feature. We applied DeepBiomarker2 to analyze the EMR data of 38,807 patients from the University of Pittsburgh Medical Center diagnosed with post-traumatic stress disorder (PTSD) to determine their risk of developing alcohol and substance use disorder (ASUD). DeepBiomarker2 predicted whether a PTSD patient would have a diagnosis of ASUD within the following 3 months with an average c-statistic (receiver operating characteristic AUC) of 0.93 and average F1 score, precision, and recall of 0.880, 0.895, and 0.866 in the test sets, respectively. Our study found that the medications clindamycin, enalapril, penicillin, valacyclovir, Xarelto/rivaroxaban, moxifloxacin, and atropine and the SDoH parameters access to psychotherapy, living in zip codes with a high normalized vegetative index, Gini index, and low-income segregation may have potential to reduce the risk of ASUDs in PTSD. In conclusion, the integration of SDoH information, coupled with the refined feature contribution analysis, empowers DeepBiomarker2 to accurately predict ASUD risk. Moreover, the model can further identify potential indicators of increased risk along with medications with beneficial effects.

Prediction of Adverse Events Risk in Patients with Comorbid Post- Traumatic Stress Disorder and Alcohol Use Disorder Using Electronic Medical Records by Deep Learning Models.

Background: Prediction of high-risk events in mental disorder patients is crucial. In our previous study, we developed a deep learning model: DeepBiomarker by using electronic medical records (EMR) to predict suicide related event (SRE) risk in post-traumatic stress disorder (PTSD) patients. Methods: We applied DeepBiomarker2 through data integration of multimodal information: lab test, medication, co-morbidities, and social determinants of health. We analyzed EMRs of 5,565 patients from University of Pittsburgh Medical Center with a diagnosis of PTSD and alcohol use disorder (AUD) on risk of developing an adverse event (opioid use disorder, SREs, depression and death). Results: DeepBiomarker2 predicted whether a PTSD + AUD patient will have a diagnosis of any adverse events (SREs, opioid use disorder, depression, death) within 3 months with area under the receiver operator curve (AUROC) of 0.94. We found piroxicam, vilazodone, dronabinol, tenofovir, suvorexant, empagliflozin, famciclovir, veramyst, amantadine, sulfasalazine, and lamivudine to have potential to reduce risk. Conclusions: DeepBiomarker2 can predict multiple adverse event risk with high accuracy and identify potential risk and beneficial factors. Our results offer suggestions for personalized interventions in a variety of clinical and diverse populations.

The impacts and mechanisms of dietary proteins on glucose homeostasis and food intake: a pivotal role of gut hormones.

Glucose and energy metabolism disorders are the main reasons induced type 2 diabetes (T2D) and obesity. Besides providing energy, dietary nutrients could regulate glucose homeostasis and food intake via intestinal nutrient sensing induced gut hormone secretion. However, reviews regarding intestinal protein sensing are very limited, and no accurate information is available on their underlying mechanisms. Through intestinal protein sensing, dietary proteins regulate glucose homeostasis and food intake by secreting gut hormones, such as glucagon-like peptide 1 (GLP-1), cholecystokinin (CCK), peptide YY (PYY) and glucose-dependent insulinotropic polypeptide (GIP). After activating the sensory receptors, such as calcium-sensing receptor (CaSR), peptide transporter-1 (PepT1), and taste 1 receptors (T1Rs), protein digests induced Ca2+ influx and thus triggered gut hormone release. Additionally, research models used to study intestinal protein sensing have been emphasized, especially several innovative models with excellent physiological relevance, such as co-culture cell models, intestinal organoids, and gut-on-a-chips. Lastly, protein-based dietary strategies that stimulate gut hormone secretion and inhibit gut hormone degradation are proposed for regulating glucose homeostasis and food intake.

Evaluating the In Situ Insulinotropic Effects of Pea Protein Hydrolysates Mediated by Active GLP-1 via a 2D and Dual-Layered Coculture Cell Model.

The aim of this study was to evaluate the in situ insulinotropic effects of pea protein hydrolysates (PPHs) mediated by active glucagon-like peptide-17-36 (active GLP-1) using a 2D and dual-layered coculture cell model. Following this model, a mixed Caco-2 and NCI-H716 cell monolayer was differentiated on the apical side to study the effects of PPHs on active GLP-1 levels; meanwhile, the beta-TC-6 cells were seeded on the basolateral side to investigate the insulin responses induced by active GLP-1. The in situ DPP-4 half-maximal inhibitory concentration (IC50) of PPHs, PPHs-120G, and PPHs-120I was 2.94, 3.43, and 2.26 mg/mL, respectively. They directly stimulated active GLP-1 secretion in NCI-H716 cells by 3.03 ± 0.21, 1.99 ± 0.03, and 2.24 ± 0.02 times, respectively. Insulin release in beta-TC-6 cells was directly stimulated by PPHs but not by PPHs-120G and PPHs-120I. Interestingly, PPHs-120G and PPHs-120I indirectly stimulated insulin release in this coculture cell model by enhancing active GLP-1 concentrations. More importantly, PPHs, PPHs-120G, and PPHs-120I increase active GLP-1 levels by their dual function of stimulating active GLP-1 secretion and DPP-4 inhibition. This study suggests that the 2D and dual-layered coculture cell model supports a more comprehensive assessment of in situ insulinotropic effects of protein hydrolysates mediated by active GLP-1.

Transepithelial Transport of the Bifunctional Peptide IPYWTY Indirectly Induced Insulin Release Mediated by Active GLP-1.

There is currently no appropriate cell model suitable for evaluating the insulinotropic effects of DPP-4 inhibitory peptides (DPP-4IPs) mediated by active glucagon-like peptide-17-36 (active GLP-1). The study aims to evaluate the transepithelial transport of IPYWTY on its in situ insulinotropic effects by using a 2D and dual-layered coculture cell model that consists of Caco-2 and NCI-H716 cells on the apical (AP) side and ß-TC-6 cells on the basolateral (BL) side. During transportation, IPYWTY was absorbed in its intact form through PepT1 and paracellular transport. Meanwhile, it was degraded to several peptide fragments, including PYWTY, YWTY, WTY, and IPY, which decreased its in situ DPP-4 inhibitory activity. IPYWTY does not directly stimulate insulin release in ß-TC-6 cells, while it increased the active GLP-1 level from 76.57 ± 15.16 to 95.63 ± 1.99 pM (1.25 times) in NCI-H716 cells. Interestingly, IPYWTY indirectly increased insulin levels from 426.91 ± 6.07 to 573.94 ± 2.97 µIU/mL (1.34 times) in the 2D and dual-layered coculture cell model for its dual function of stimulating active GLP-1 secretion and DPP-4 inhibition. These results suggested that the 2D and dual-layered coculture cell model is an alternative strategy for effectively evaluating the insulinotropic effects of DPP-4IPs mediated by active GLP-1.

Unveiling the Enigma: Exploring Risk Factors and Mechanisms for Psychotic Symptoms in Alzheimer's Disease through Electronic Medical Records with Deep Learning Models.

Around 50% of patients with Alzheimer's disease (AD) may experience psychotic symptoms after onset, resulting in a subtype of AD known as psychosis in AD (AD + P). This subtype is characterized by more rapid cognitive decline compared to AD patients without psychosis. Therefore, there is a great need to identify risk factors for the development of AD + P and explore potential treatment options. In this study, we enhanced our deep learning model, DeepBiomarker, to predict the onset of psychosis in AD utilizing data from electronic medical records (EMRs). The model demonstrated superior predictive capacity with an AUC (area under curve) of 0.907, significantly surpassing conventional risk prediction models. Utilizing a perturbation-based method, we identified key features from multiple medications, comorbidities, and abnormal laboratory tests, which notably influenced the prediction outcomes. Our findings demonstrated substantial agreement with existing studies, underscoring the vital role of metabolic syndrome, inflammation, and liver function pathways in AD + P. Importantly, the DeepBiomarker model not only offers a precise prediction of AD + P onset but also provides mechanistic understanding, potentially informing the development of innovative treatments. With additional validation, this approach could significantly contribute to early detection and prevention strategies for AD + P, thereby improving patient outcomes and quality of life.

DeepBiomarker2: Prediction of alcohol and substance use disorder risk in post-traumatic stress disorder patients using electronic medical records and multiple social determinants of health.

Introduction: Prediction of high-risk events amongst patients with mental disorders is critical for personalized interventions. In our previous study, we developed a deep learning-based model, DeepBiomarker by utilizing electronic medical records (EMR) to predict the outcomes of patients with suicide-related events in post-traumatic stress disorder (PTSD) patients. Methods: We improved our deep learning model to develop DeepBiomarker2 through data integration of multimodal information: lab tests, medication use, diagnosis, and social determinants of health (SDoH) parameters (both individual and neighborhood level) from EMR data for outcome prediction. We further refined our contribution analysis for identifying key factors. We applied DeepBiomarker2 to analyze EMR data of 38,807 patients from University of Pittsburgh Medical Center diagnosed with PTSD to determine their risk of developing alcohol and substance use disorder (ASUD). Results: DeepBiomarker2 predicted whether a PTSD patient will have a diagnosis of ASUD within the following 3 months with a c-statistic (receiver operating characteristic AUC) of 0·93. We used contribution analysis technology to identify key lab tests, medication use and diagnosis for ASUD prediction. These identified factors imply that the regulation of the energy metabolism, blood circulation, inflammation, and microbiome is involved in shaping the pathophysiological pathways promoting ASUD risks in PTSD patients. Our study found protective medications such as oxybutynin, magnesium oxide, clindamycin, cetirizine, montelukast and venlafaxine all have a potential to reduce risk of ASUDs. Discussion: DeepBiomarker2 can predict ASUD risk with high accuracy and can further identify potential risk factors along with medications with beneficial effects. We believe that our approach will help in personalized interventions of PTSD for a variety of clinical scenarios.

Effect of fiber-bound polyphenols from highland barley on lipid oxidation products of cooked pork during in vitro gastrointestinal digestion.

BACKGROUND: The gastrointestinal (GI) tract is a major site of lipid oxidation, and the lipid oxidation products are related to an increased risk of various chronic diseases. In this study, the inhibition capacity of bound-polyphenol rich insoluble dietary fiber (BP-IDF) from highland barley (HB) to lipid oxidation was evaluated during simulated GI digestion. RESULTS: We found that the level of lipid hydroperoxides (LOOH) and aldehydes were significantly inhibited when highland barley bound-polyphenol rich insoluble dietary fiber (HBBP-IDF) co-digestion with cooked pork. The lipid oxidation products were more effectively scavenged during simulated gastric digestion, with inhibition of 77.4% for LOOH, 52.3% for malondialdehyde, 46.5% for 4-hydroxy-2-hexenal and 48.7% for 4-hydroxy-2-nonenel, respectively. The fiber-bound polyphenols are the principal scavengers of lipid oxidation products. CONCLUSION: These findings suggest that HBBP-IDF could be used as a functional ingredient able to scavenge lipid oxidation products across the GI tract. © 2023 Society of Chemical Industry.

A systematic review and meta-analysis: Effects of protein hydrolysate supplementation on fat-free mass and strength in resistance-trained individuals.

The quality of the existing evidence on the effects of protein hydrolysate supplementation on fat-free mass (FFM) and upper and lower body strength under resistance exercise intervention has not been evaluated. We conducted a structured literature search in PubMed, Web of Science, Cochrane Library, and Scopus database. A random effect model was used with continuous data of FFM and upper and lower body strength for healthy participants over 18 years old who received resistance training for ≥4 weeks and took protein hydrolysate or equivalent control supplements. Sensitivity and subgroup analyses were also conducted. Data from 330 participants in eight studies showed that supplemental protein hydrolysate had a positive effect on the FFM (n = 13, SMD = 0.36, 95% confidence interval (CI): 0.16-0.56, P = 0.000) and lower (n = 7, SMD = 0.43, 95% CI: 0.16-0.69, P = 0.001) and upper (n = 5, SMD = 0.17, 95% CI: -0.06-0.41, P = 0.145) body strength of resistance-trained individuals compared with placebo, showing an increase in physical fitness and muscle strength. However, the current evidence is insufficient to establish ingestion recommendations.

Effect of cyclodextrin glucosyltransferase extracted from Bacillus xiaoxiensis on wheat dough and bread properties.

In this study, the cyclodextrin glucosyltransferase (CGTase) was extracted from Bacillus xiaoxiensis. CGTase had negative effects on dough viscoelastic properties and gluten strength but had positive effects on bread baking qualities and anti-staling properties. Adding an appropriate amount of CGTase (less than 0.3 U/g) could improve the specific volume, crumb texture, crust color, moisture content, and crumb hardness of bread. The bread crumb with 0.4 U/g CGTase (based on flour weight) had the lowest retrogradation enthalpy of 0.53 ± 0.10 J/g and the lowest relative crystallinity of 16.1%, which indicated the alleviating effect of amylopectin crystallization. Moreover, CGTase reduced the moisture from forming crystal lattices and limited starch molecule migration. The T2 transverse relaxation results showed that the increase of immobilized water content in the bread with CGTase was lower than the control after 5 days of storage, which implied the water-holding capacity of the bread was enhanced and provided information on the inhibition of water migration. Hence, the CGTase could be a potential bread improver.

Food-derived dipeptidyl peptidase IV inhibitory peptides: Production, identification, structure-activity relationship, and their potential role in glycemic regulation.

Dipeptidyl Peptidase IV (DPP-IV) inhibitory peptides are attracting increasing attention, owing to their potential role in glycemic regulation by preventing the inactivation of incretins. However, few reviews have summarized the current understanding of DPP-IV inhibitory peptides and their knowledge gaps. This paper reviews the production, identification and structure-activity relationships (SAR) of DPP-IV inhibitory peptides. Importantly, their bioavailability and hypoglycemic effects are critically discussed. Unlike the traditional method to identifying peptides after separation step by step, the bioinformatics approach identifies peptides via virtual screening that is more convenient and efficient. In addition, the bioinformatics approach was also used to investigate the SAR of peptides. Peptides with proline (Pro) or alanine (Ala) residue at the second position of N-terminal are exhibit strong DPP-IV inhibitory activity. Besides, the bioavailability of DPP-IV inhibitory peptides is related to their gastrointestinal stability and cellular permeability, and in vivo studies showed that the glucose homeostasis has been improved by these peptides. Especially, the intestinal transport of DPP-IV inhibitory peptides and cell biological assays used to evaluate their potential role in glycemic regulation are innovatively summarized. For further successful development of DPP-IV inhibitory peptides in glycemic regulation, future study should elucidate their SAR and in vivo hypoglycemic effects .

Rapid Screening of Novel Dipeptidyl Peptidase-4 Inhibitory Peptides from Pea (Pisum sativum L.) Protein Using Peptidomics and Molecular Docking.

Pea protein hydrolysates (PPHs) possess good hypoglycemic effects; however, their dipeptidyl peptidase-4 (DPP-4) inhibitory activity is poorly understood, and none of the DPP-4 inhibitory peptides have been identified from PPHs. This paper aims to rapidly screen these peptides from PPHs by combining peptidomics and molecular docking. In this study, 543 peptides were identified by peptidomics, and four peptides (IPYWTY, IPYWT, LPNYN, and LAFPGSS) with DPP-4 half-maximal inhibitory concentration (IC50) values <100 µM were screened for the first time. Significantly, peptide IPYWTY exhibited the most potent DPP-4 inhibitory activity (IC50 = 11.04 µM) mainly because it formed hydrophobic interactions with the S1 pocket in DPP-4. These results indicated that combining peptidomics and molecular docking is an effective strategy for rapidly screening DPP-4 inhibitory peptides.

Trapping of reactive carbonyl species by fiber-bound polyphenols from whole grains under simulated physiological conditions.

The carbonyl trapping activity of bound-polyphenol rich insoluble dietary fiber (BP-IDF) from different whole grains and underlying mechanism of these BP-IDF actions were studied under simulated physiological conditions. We found that the black highland barley BP-IDF exhibited the most pronounced effect in scavenging carbonyls by trapping 88.7%, 72.2%, 95.7%, and 31.4% for methylglyoxal, glyoxal, acrolein, and malondialdehyde within 24 h, respectively. After vitro gastrointestinal digestion, the black highland barley BP-IDF still retained considerable trapping activity for carbonyls. The carbonyl scavenging capacity was reduced by up to 93% after removing bound polyphenols from the black highland barley BP-IDF, which was consistent with the reduction in its total phenolic content. Moreover, the formation of adducts between reactive carbonyl species (RCS) and polyphenols bound to insoluble dietary fiber (IDF) was also detected. Overall, these findings confirmed that IDF-bound polyphenols were still active to trap RCS, indicating the potential benefits of BP-IDF from whole grains as functional ingredients to limit carbonyl stress across the gastrointestinal tract.

The Influence of Water-Unextractable Arabinoxylan and Its Hydrolysates on the Aggregation and Structure of Gluten Proteins.

This study aimed to investigate the influence of water-unextractable arabinoxylan (WUAX) and its hydrolysates on the aggregation and structure of gluten proteins and reveal the underlying mechanism. In this work, the WUAX was treated with enzymatic hydrolysis and the changes of their molecular weights and structures were analyzed. Meanwhile, the conformation and aggregation of gluten were determined by reversed-phase HPLC, FT-Raman spectroscopy, and confocal laser scanning microscopy. The results showed that the extra WUAX could impair the formation of high Mw glutenin subunits, and the enzymatic hydrolysis arabinoxylan (EAX) could induce the aggregation of gluten subunits. And, the gluten microstructure was destroyed by WUAX and improved by EAX. Besides, the interactions of WUAX and EAX with gluten molecules were different. In summary, these results indicated that enzymatic hydrolysis changed the physicochemical properties of arabinoxylan and affected the interaction between arabinoxylan and gluten proteins.

DeepBiomarker: Identifying Important Lab Tests from Electronic Medical Records for the Prediction of Suicide-Related Events among PTSD Patients.

Identifying patients with high risk of suicide is critical for suicide prevention. We examined lab tests together with medication use and diagnosis from electronic medical records (EMR) data for prediction of suicide-related events (SREs; suicidal ideations, attempts and deaths) in post-traumatic stress disorder (PTSD) patients, a population with a high risk of suicide. We developed DeepBiomarker, a deep-learning model through augmenting the data, including lab tests, and integrating contribution analysis for key factor identification. We applied DeepBiomarker to analyze EMR data of 38,807 PTSD patients from the University of Pittsburgh Medical Center. Our model predicted whether a patient would have an SRE within the following 3 months with an area under curve score of 0.930. Through contribution analysis, we identified important lab tests for suicide prediction. These identified factors imply that the regulation of the immune system, respiratory system, cardiovascular system, and gut microbiome were involved in shaping the pathophysiological pathways promoting depression and suicidal risks in PTSD patients. Our results showed that abnormal lab tests combined with medication use and diagnosis could facilitate predicting SRE risk. Moreover, this may imply beneficial effects for suicide prevention by treating comorbidities associated with these biomarkers.

Investigation the influences of water-extractable and water-unextractable arabinoxylan on the quality of whole wheat you-tiao and its mechanism.

This research aimed to investigate the effects of water-extractable arabinoxylan (WEAX) and water-unextractable arabinoxylan (WUAX) on the quality of you-tiao. In this work, the interactions between different amounts of AX and wheat gluten were extensively evaluated during frying treatment. The results showed that WEAX impaired the surface hydrophobicity of gluten and improved its solubility, while WUAX had the opposite effect. The fluorescence spectra revealed that WEAX and WUAX changed the conformation of gluten molecules. Besides, chemical interaction measurement indicated that WEAX and WUAX prevented the formation of partial disulfide bonds and inhibited the thermal aggregation of gluten proteins. In summary, the results indicated that WEAX partly improved the properties of you-tiao. Meanwhile, WUAX reduced the dough's oil content and specific volume, resulting in you-tiao with poor quality.