PubMed Health⌕ Search

Biomedical subjects

Xiaoying Xu

Publications and source records attributed to Xiaoying Xu.

10 recordsLinked to original sources
PubMed2026

A bimodal large language model reduces misalignment in patient education: A double-blinded randomized trial.

BACKGROUND: Effective patient education requires accurate communication aligned with patients' emotional and semantical needs. Text-based large language models (LLMs) lack access to non-verbal cues, which may contribute to misaligned responses. METHODS: We evaluated emotional and semantic misalignment in a text-based LLM using 64,200 utterances from 16,583 patient education cases across six departments and three centers. Dolphin was developed integrating text and audio cues and evaluated through emotion recognition, semantic consistency assessment, branch-level ablations, and a double-blinded randomized trial against a matched text-based LLM comparator (Chinese Clinical Trial Registry: (ChiCTR2500095933). FINDINGS: The text-based LLM showed emotional misalignment in 36.7% of responses and semantic misalignment in 28.3% of cases, with higher misalignment under greater burden. Dolphin outperformed the text-based LLM in emotion recognition accuracy (0.886 vs. 0.713) and semantic consistency (84.9% vs. 82.1%; both adjusted p < 0.001). Ablations supported contribution of audio branches. Dolphin received higher expert ratings than the text-based LLM and human educators (all p < 0.001). In 555 patients, Dolphin was associated with greater patient satisfaction (98.6% vs. 93.8%), suggestion acceptance (76.1% vs. 58.9%; p < 0.001), proactive disclosure (44.6% vs. 26.5%; p < 0.001), and fewer 7-day unplanned recontact (12.9% vs. 22.9%; p = 0.002). No unsafe recommendations or safety events were identified. CONCLUSIONS: Compared with text-based LLM, Dolphin improved emotional-semantic alignment and patient-education outcomes, supporting bimodal alignment as a strategy for reducing misalignment-driven communication failures. FUNDING: National Natural Science Foundation of China, State Key Laboratory Special Fund, and Chinese Academy of Medical Sciences Innovation Fund.

Humans↗
PubMed2026

Cis-regulatory evolution of CsANS1 drives cultivar variation in anthocyanin accumulation in tea plants.

Anthocyanins, a ubiquitous class of water-soluble phytochemicals renowned for their chromatic diversity and potent bioactivity, are integral to the phenotypic and metabolic plasticity of higher plants. Using an integrative multi-omics approach that combines transcriptomic and metabolomic profiling, we identified anthocyanin synthase (CsANS1) as the key genetic determinant responsible for interspecific variation in anthocyanin accumulation among tea plants. Architectural comparison of promoter regions revealed a 192-bp variation insertion in the CsANS1 cis-regulatory region with potential functional significance. This insertion was strictly conserved in anthocyanin-rich (purple-leaf) cultivars, including both natural and hybrid genotypes, but entirely missing in anthocyanin-deficient (green-leaf) cultivars. Dual-luciferase assays confirmed that this insertion enhances promoter activity. Additionally, we delineated a tripartite regulatory axis comprising CsmiR156b, CsSPL9, and CsMYB75 which orchestrates the spatiotemporal modulation of CsANS1 expression and, consequently, anthocyanin biosynthesis. Collectively, these findings provide a mechanistic paradigm for anthocyanin polymorphism in tea plants, implicating both cis-regulatory evolution and transcriptional network synergy as pivotal drivers of phytochemical diversification.

Anthocyanins↗
PubMed2005

The conserved residue Phe273(282) of PPARalpha(gamma), beyond the ligand-binding site, functions in binding affinity through solvation effect.

Peroxisome proliferator-activated receptors (PPARs) belong to the members of the nuclear receptor superfamily, and play important roles in lipid and glucose homeostasis. Residue Phe282 in PPARgamma (Phe273 in PPARalpha), beyond the ligand-binding site, is a conserved amino acid across several nuclear receptors and in all PPAR subfamily. In this work, we firstly investigated the influence of Phe282(273)Ala mutation on the binding affinity of PPARgamma(alpha) against a series of agonists by use of surface plasmon resonance (SPR) technique and cellular transcriptional activation analysis. Phe282(273)Ala mutation decreases the binding affinities of the ligands to the receptors in certain degrees, from several to 1000-folds. Phe282Ala mutation dramatically reduced the binding affinity of PPARgamma to GI262570, however, this mutation did not affect PPARalpha binding to this ligand, thereby suggesting that the Phe282 and Phe273 are associated with the selectivity of GI262570 binding to PPARgamma and PPARalpha. The mutation reduced the transcriptional activation activities of the receptors induced by the ligand binding, and the decrease degree is generally in agreement with the binding affinities of the ligands to the receptors. The 5 ns MD simulations for the wild-type and mutated PPARgamma showed that the mutation did not influence the flexibility of the receptor. There is no repulsion between Phe282 and the proceeding loop of AF2. However, substitution of Phe282 by alanine enlarged the entrance of the binding pocket and abolished the repulsive interaction between solvent water molecules and this hydrophobic residue, thus more water molecules can enter into the binding pocket. It needs more energy to exclude the extra water molecules for a ligand binding to the mutated receptor. In addition, the extra water molecules abolish some of H-bonds between the ligand and receptors. Therefore, solvent effect may be concluded as the major source of the decrease of binding affinity for the mutated receptors to the ligands, and thereby of the decrease of their transcriptional activation activities.

Amino Acid Sequence↗
PubMed2005

Development of a low volume plasma sample precipitation procedure for liquid chromatography/tandem mass spectrometry assays used for drug discovery applications.

The demand for high sensitivity bioanalytical methods has dramatically increased in the drug discovery stage; in addition, there has been a growing trend of reducing the sample volume that is required for these assays. A sensitive high-performance liquid chromatography/tandem mass spectrometry (HPLC/MS/MS) procedure has been developed and tested to meet these needs. The assay requires only a low plasma sample volume (10 microL) and employs a protein precipitation procedure using a 1:6 plasma/acetonitrile ratio. The supernatant is injected directly into the LC/MS/MS system using the selected reaction monitoring (SRM) procedure for detection. A generic HPLC gradient based on a methanol/water mobile phase with a flow rate set to 0.8 mL/min was used. The test method showed very good linearity between 0.1-1000 ng/mL (R2 = 0.9737), precision (%RSD = 6-9), accuracy (%RE = -2) and reproducibility (%RSD = 11). A drug discovery IV/PO study was assayed using both the new low volume method and our standard volume (50 microL) method. The correlation of the two sets of data from the two methods was excellent (R2 = 0.9287). This new assay procedure has been successfully used in our laboratory for over 100 different rat or mouse discovery PK studies.

Animals↗
PubMed2005

A study of common discovery dosing formulation components and their potential for causing time-dependent matrix effects in high-performance liquid chromatography tandem mass spectrometry assays.

Hydroxyproyl-beta-cyclodextran (HPBCD), methyl cellulose (MC), Tween 80 and PEG400 are commonly used in dosing formulations in pharmacokinetic (PK) studies during the early drug discovery stage. A series of studies was designed to evaluate the potential matrix effects of these dosing vehicles when the samples are assayed by high-performance liquid chromatography combined with tandem mass spectrometry (HPLC/MS/MS). These dosing vehicles were dosed into the rats via either an intravenous (IV) or an oral route (PO) and plasma samples were collected for a 24-h post-dose period. Five test compounds with CLog P values ranging from 0.9 to 5.4 were spiked into the collected rat plasma. After protein precipitation, these samples were analyzed using a generic fast-gradient HPLC/MS/MS method. Three popular mass spectrometers (Thermo-Finnigan Quantum with ESI and APCI, AB-Sciex API 3000 with ESI and APCI, and Waters-Micromass Quattro Ultima with ESI) were used to test these plasma samples. Results indicated that there was no observed matrix effect for all five compounds when 20% HPBCD or 0.4% MC was used as the vehicle in either the IV or the PO route, respectively. In addition, 0.1% Tween 80 dosed either IV or PO caused significant ion suppression (50-80%, compared to results obtained from plasma samples free from vehicles) for compounds that eluted at the beginning of the chromatogram. Also, PEG400 when used in an oral formulation caused significant ion suppression (30-50%) for early eluting compounds. These matrix effects were not only ionization mode (ESI or APCI) dependent, but also source design (Thermo-Finnigan, AB-Sciex or Waters-Micromass) dependent. Overall, the APCI mode proved to be less vulnerable to matrix effects than the ESI mode. Some possible mechanisms of these matrix effects are proposed and simple strategies to avoid these matrix effects are discussed.

Animals↗
PubMed2003

Investigation of matrix effects in bioanalytical high-performance liquid chromatography/tandem mass spectrometric assays: application to drug discovery.

A series of studies was performed to investigate some of the causes for matrix effects ('ion suppression' or 'ion enhancement') in bioanalytical high-performance liquid chromatography/tandem mass spectrometry (HPLC/MS/MS) assays. Previous studies have reported that matrix effects are mainly due to endogenous components in biological fluids and are a greater concern for electrospray ionization (ESI) than for atmospheric pressure chemical ionization (APCI). In this report we demonstrate that: (1) matrix effects can also be caused by exogenous materials, such as polymers contained in different brands of plastic tubes, or Li-heparin, a commonly used anticoagulant; (2) matrix effects are not only ionization mode (APCI or ESI) dependent, but also source design (Sciex, Finnigan, Micromass) dependent; and (3) for at least one vendor's design, we found the APCI mode to be more sensitive to matrix effects than the ESI mode. Based on these findings, we have proposed the following simple strategies to avoid matrix effects: (1) select the same brand of plastic tubes for processing and storing plasma samples and spiked plasma standards; (2) avoid using Li-heparin as the anticoagulant; and (3) try switching the ionization mode or switching to different mass spectrometers when matrix effects are encountered. These three strategies have allowed us to use protein precipitation and generic fast LC techniques to generate reliable LC/MS/MS data for the support of pharmacokinetic studies at the early drug discovery stage.

Animals↗
PubMed2003

Comparison of conventional and enhanced mass resolution triple-quadrupole mass spectrometers for discovery bioanalytical applications.

A new triple-quadrupole mass spectrometry (MS) system with enhanced resolution capabilities has recently become available. In order to evaluate the performance of this new instrument for drug discovery assays, both the linearity and the limit of detection were compared in the positive electrospray ionization (ESI) mode with those of a conventional triple-quadrupole instrument supplied by the same manufacturer. For these studies, spiked mouse plasma standard samples were split and assayed by each instrument, which allowed for a direct comparison of the two systems. In the unit mass resolution mode, the new mass spectrometer was found to be at least ten-fold more sensitive than the conventional instrument. The sensitivity of the new mass spectrometer under the enhanced mass resolution mode was found to be even better by another factor of two. For the test compound, the linear dynamic range was found to be 0.05-5000 ng/mL for the new instrument as compared with 2.5-5000 ng/mL for the conventional mass spectrometer.

Animals↗
PubMed2003

Virtual screening on natural products for discovering active compounds and target information.

Natural products, containing inherently large-scale structural diversity than synthetic compounds, have been the major resources of bioactive agents and will continually play as protagonists for discovering new drugs. However, how to access this diverse chemical space efficiently and effectively is an exciting challenge for medicinal chemists and pharmacologists. While virtual screening, which has shown a great promise in drug discovery, will play an important role in digging out lead (active) compounds from natural products. This review focuses on the strategy of virtual screening based on molecular docking and, with successful examples from our laboratory, illustrates the efficiency of virtual screening in discovering active compounds from natural products. On the other hand, the sequencing of the human genome and numerous pathogen genomes has resulted in an unprecedented opportunity for discovering potential new drug targets. Chemogenomics has appeared as a new technology to initiate target discovery by using active compounds as probes to characterize proteome functions. Natural products are the ideal probes for such research. Binding affinity fingerprint is a powerful chemogenomic descriptor to characterize both small molecules and pharmacologically relevant proteins. Therefore, this review also discusses binding affinity fingerprint strategy for identifying target information from the genomic data by using natural products as the probes.

Biological Products↗
PubMed2002

Liquid chromatography-mass spectrometry of cis- and all-trans-lycopene in human serum and prostate tissue after dietary supplementation with tomato sauce.

Several epidemiological studies suggest a lower incidence of prostate cancer in men who routinely consume tomato products. Tomatoes are the primary dietary source of lycopene, which is among the most potent antioxidants of the carotenoids. Men with clinical stage T1 or T2 prostate adenocarcinoma were recruited (n = 32) and consumed tomato sauce based pasta dishes for 3 weeks (equivalent to 30 mg of lycopene per day) before radical prostectomy. Prostate tissue from needle biopsy just before intervention and prostectomy after supplementation from a subset of 11 subjects was evaluated for both total lycopene and lycopene geometrical isomer ratios. A gradient HPLC system using a C(18) column with UV-vis absorbance detection was used to measure total lycopene. Because the absorbance detector was insufficiently sensitive, HPLC with a C(30) column and positive ion atmospheric pressure chemical ionization mass spectrometric (LC-MS) detection was developed as a new assay to measure the ratio of lycopene cis/trans isomers in these samples. The limit of detection of the LC-MS method was determined to be 0.93 pmol of lycopene on-column, and a linear response was obtained over 3 orders of magnitude. Total lycopene in serum increased 2.0-fold from 35.6 to 69.9 microg/dL (from 0.664 to 1.30 microM) as a result of dietary supplementation with tomato sauce, whereas total lycopene in prostate tissue increased 3.0-fold from 0.196 to 0.582 ng/mg of tissue (from 0.365 to 1.09 pmol/mg). all-trans-Lycopene and at least 14 cis-isomer peaks were detected in prostate tissue and serum. The mean proportion of all-trans-lycopene in prostate tissue was approximately 12.4% of total lycopene before supplementation but increased to 22.7% after dietary intervention with tomato sauce. In serum there was only a 2.8% but statistically significant increase in the proportion of all-trans-lycopene after intervention. These results indicate that short-term supplementation with tomato sauce containing primarily all-trans-lycopene (83% of total lycopene) results in substantial increases in total lycopene in serum and prostate and a substantial increase in all-trans-lycopene in prostate but relatively less in serum.

Adenocarcinoma↗