Biomedical subjects

Yixin Chen

Publications and source records attributed to Yixin Chen.

18 recordsLinked to original sources

PubMed2026

Primer design through submodular function estimation.

MOTIVATION: Multiplex PCR-based enrichment is widely used in viral genome sequencing and pathogen surveillance. However, designing large sets of primers that maximize genome coverage while minimizing primer-primer interactions remains a major computational challenge. Existing methods such as SADDLE and Olivar use heuristics to optimize a Badness score for primer dimers but lack theoretical guarantees on solution quality. RESULTS: We introduce PRISM, a new framework that formulates multiplex primer design as a constrained submodular maximization problem. Our method defines an objective that balances genome coverage and dimer risk, and applies a local search algorithm with a constant-factor approximation guarantee. Evaluations on viral genome datasets demonstrate that PRISM consistently achieves lower Badness scores compared to PrimalScheme, Olivar, and primerJinn. These results highlight the scalability and theoretical rigor of submodular optimization in primer design. AVAILABILITY: PRISM is open-source and available at https://github.com/yhhan19/PRISM-new. The experimental data, scripts, and results used in this paper are archived on Figshare at https://doi.org/10.6084/m9.figshare.32806499.

Algorithms↗

PubMed2026

BioMedGraphica: an all-in-one platform for joint textual biomedical prior knowledge and numeric graph generation.

MOTIVATION: Multiomics data analysis is essential for scientific discovery in precision medicine. However, translating analysis results of omics data analysis into novel scientific hypotheses remains a significant challenge. Human experts must manually review analysis results and generate new hypotheses based on extensive and interconnected biomedical prior knowledge, which is subjective and not scalable. While large language models can accelerate the discovery, their reasoning improves when grounded in structured, auditable, and comprehensive biomedical prior knowledge. However, biomedical knowledge is scattered across heterogeneous databases that use diverse and inconsistent nomenclature systems, making it difficult to integrate resources into a unified format for scalable analysis. This fragmentation limits the ability of artificial intelligence systems to fully leverage biomedical data for scientific discovery. RESULTS: We developed BioMedGraphica, a novel all-in-one platform that harmonizes fragmented biomedical resources by integrating 11 entity types and 30 relation types from 43 databases into a unified textual prior knowledge graph containing 2 306 921 entities and 27 232 091 relations. In addition, we present a novel textual-numeric graph (TNG) data structure concept, where textual information captures prior biological knowledge (e.g. transcription start sites, functions, mechanisms), numeric values represent quantitative biomedical features, and the integrated relations can help uncover mechanisms. By bridging prior knowledge with user-specific data, TNG is a novel and ideal data structure for developing novel graph analysis models. AVAILABILITY AND IMPLEMENTATION: The code is available at: https://github.com/FuhaiLiAiLab/BioMedGraphica and BioMedGraphica knowledge graph database can be downloaded from huggingface dataset: https://huggingface.co/datasets/FuhaiLiAiLab/BioMedGraphica.

Humans↗

PubMed2026

Safety, tolerability, and efficacy of RIPK1 inhibitor, SAR443820, in amyotrophic lateral sclerosis (HIMALAYA): a multicentre, randomised, double-blind, placebo-controlled, phase 2 trial.

BACKGROUND: RIPK1, a protein regulating inflammatory signalling and cell death, is implicated in amyotrophic lateral sclerosis (ALS) pathophysiology. SAR443820 is a selective, oral, CNS-penetrant, reversible RIPK1 inhibitor. We aimed to evaluate the safety, tolerability, and efficacy of SAR443820 in participants with ALS. METHODS: This multicentre, randomised, double-blind, placebo-controlled, phase 2 trial was conducted at 63 clinical sites in 13 countries (Belgium, Canada, China, France, Germany, Italy, Japan, the Netherlands, Poland, Spain, Sweden, the UK, and the USA). Adults (aged 18-80 years) with a diagnosis of possible ALS, clinically probable ALS, clinically probable laboratory-supported ALS, or clinically definite ALS, in accordance with the revised El Escorial World Federation of Neurology criteria, were randomly assigned (2:1) by use of a stratified block design (blocks of three) to receive either 20 mg SAR443820 orally twice per day or matching placebo in the 24-week double-blind period. Randomisation was done centrally using interactive response technology and stratified by geographical region of trial site, region of ALS onset, use of riluzole, use of edaravone, and use of the combination of sodium phenylbutyrate and taurursodiol. Participants, care providers, investigators, and outcomes assessors were masked to trial intervention. The primary outcome was a change in ALS Functional Rating Scale Revised (ALSFRS-R) total score from baseline to week 24 and was calculated for all participants who had an ALSFRS-R total score available at baseline and at week 24. Safety analyses included all randomly assigned participants receiving one dose or more of trial intervention. This trial is registered with ClinicalTrials.gov (NCT05237284) and was terminated early. FINDINGS: Between April 13, 2022, and July 17, 2023, 397 participants were screened and 305 randomly assigned to SAR443820 (n=203) or placebo (n=102); six were excluded from the primary analysis due to missing baseline ALSFRS-R values. Mean age was 56·9 years (SD 11·5); 183 (60%) participants were male and 122 (40%) were female. Least squares mean change in ALSFRS-R from baseline to week 24 was -6·73 (95% CI -7·48 to -5·98) for SAR443820 group (n=169) and -6·32 (-7·36 to -5·27) for placebo group (n=87). There was no statistically significant difference between the study groups (least squares mean difference -0·41 [95% CI -1·71 to 0·88]). Participants in the SAR443820 group had higher incidence of adverse events (171 [85%] of 202 vs placebo 80 [78%] of 102) and treatment discontinuations (28 [14%] of 202 vs placebo five [5%] of 102), with elevated hepatic enzymes being the most common cause. Nine deaths occurred in the double-blind period (seven [3%] of 202 in the SAR443820 group and two [2%] of 102 in the placebo group); none was attributed to SAR443820. INTERPRETATION: SAR443820 did not show clinical benefit and was associated with higher hepatic enzyme increase, indicating that further clinical development of SAR443820 in ALS is not warranted. FUNDING: Sanofi.

Humans↗

PubMed2025

BioMedGraphica: An All-in-One Platform for Joint Textual Biomedical Prior Knowledge and Numeric Graph Generation.

Multi-omic data analysis is essential for scientific discovery in precision medicine. However, translating statistical results of omic data analysis into novel scientific hypothesis remains a significant challenge. Human experts must manually review analysis results and generate new hypothesis based on extensive and inter-connected biomedical prior knowledge, which is subjective and not scalable. While large language models (LLMs) can accelerate the discovery, their reasoning improves when grounded in structured, auditable and comprehensive biomedical prior knowledge. Biomedical knowledge, however, is scattered across heterogeneous databases that use diverse and inconsistent nomenclature systems, making it difficult to integrate resources into a unified format for scalable analysis. This fragmentation limits the ability of AI systems to fully leverage biomedical data for scientific discovery. To address these challenges, we developed BioMedGraphica , an all-in-one platform that harmonizes fragmented biomedical resources by integrating 11 entity types and 30 relation types from 43 databases into a unified knowledge graph containing 2,306,921 entities and 27,232,091 relations. In addition, to the best of our knowledge, this is the first work to propose a novel Textual-Numeric Graph (TNG) data-structure for multi-omics data analysis. In TNG, textual information captures prior biological knowledge (e.g., transcription start sites, functions, mechanisms), while numeric values represent quantitative biomedical features, and the integrated relations can help uncover mechanisms. By bridging prior knowledge with user-specific data, TNG is a novel and ideal data-structure for the development of graph foundation models, with the potential to improve prediction performance and interpretability, while also augmenting LLMs by supplying graph-structured mechanistic context to strengthen reasoning. The details for BioMedGraphica code can be accessed by github link: https://github.com/FuhaiLiAiLab/BioMedGraphica and BioMedGraphica knowledge graph data can be downloaded from huggingface dataset: https://huggingface.co/datasets/FuhaiLiAiLab/BioMedGraphica.

biomedical knowledge graph↗

PubMed2006

A novel approach to phylogenetic tree construction using stochastic optimization and clustering.

BACKGROUND: The problem of inferring the evolutionary history and constructing the phylogenetic tree with high performance has become one of the major problems in computational biology. RESULTS: A new phylogenetic tree construction method from a given set of objects (proteins, species, etc.) is presented. As an extension of ant colony optimization, this method proposes an adaptive phylogenetic clustering algorithm based on a digraph to find a tree structure that defines the ancestral relationships among the given objects. CONCLUSION: Our phylogenetic tree construction method is tested to compare its results with that of the genetic algorithm (GA). Experimental results show that our algorithm converges much faster and also achieves higher quality than GA.

Algorithms↗

PubMed2006

An improved ant colony algorithm with diversified solutions based on the immune strategy.

BACKGROUND: Ant colony algorithm has emerged recently as a new meta-heuristic method, which is inspired from the behaviours of real ants for solving NP-hard problems. However, the classical ant colony algorithm also has its defects of stagnation and premature. This paper aims at remedying these problems. RESULTS: In this paper, we propose an adaptive ant colony algorithm that simulates the behaviour of biological immune system. The solutions of the problem are much more diversified than traditional ant colony algorithms. CONCLUSION: The proposed method for improving the performance of traditional ant colony algorithm takes into account the polarization of the colonies, and adaptively adjusts the distribution of the solutions obtained by the ants. This makes the solutions more diverse so as to avoid the stagnation and premature phenomena.

Algorithms↗

PubMed2006

A fast parallel algorithm for finding the longest common sequence of multiple biosequences.

BACKGROUND: Searching for the longest common sequence (LCS) of multiple biosequences is one of the most fundamental tasks in bioinformatics. In this paper, we present a parallel algorithm named FAST_LCS to speedup the computation for finding LCS. RESULTS: A fast parallel algorithm for LCS is presented. The algorithm first constructs a novel successor table to obtain all the identical pairs and their levels. It then obtains the LCS by tracing back from the identical character pairs at the last level. Effective pruning techniques are developed to significantly reduce the computational complexity. Experimental results on gene sequences in the tigr database show that our algorithm is optimal and much more efficient than other leading LCS algorithms. CONCLUSION: We have developed one of the fastest parallel LCS algorithms on an MPP parallel computing model. For two sequences X and Y with lengths n and m, respectively, the memory required is max{4*(n+1)+4*(m+1), L}, where L is the number of identical character pairs. The time complexity is O(L) for sequential execution, and O(|LCS(X, Y)|) for parallel execution, where |LCS(X, Y)| is the length of the LCS of X and Y. For n sequences X1, X2, ..., Xn, the time complexity is O(L) for sequential execution, and O(|LCS(X1, X2, ..., Xn)|) for parallel execution. Experimental results support our analysis by showing significant improvement of the proposed method over other leading LCS algorithms.

Algorithms↗

PubMed2006

MILES: multiple-instance learning via embedded instance selection.

Multiple-instance problems arise from the situations where training class labels are attached to sets of samples (named bags), instead of individual samples within each bag (called instances). Most previous multiple-instance learning (MIL) algorithms are developed based on the assumption that a bag is positive if and only if at least one of its instances is positive. Although the assumption works well in a drug activity prediction problem, it is rather restrictive for other applications, especially those in the computer vision area. We propose a learning method, MILES (Multiple-Instance Learning via Embedded instance Selection), which converts the multiple-instance learning problem to a standard supervised learning problem that does not impose the assumption relating instance labels to bag labels. MILES maps each bag into a feature space defined by the instances in the training bags via an instance similarity measure. This feature mapping often provides a large number of redundant or irrelevant features. Hence, 1-norm SVM is applied to select important features as well as construct classifiers simultaneously. We have performed extensive experiments. In comparison with other methods, MILES demonstrates competitive classification accuracy, high computation efficiency, and robustness to labeling uncertainty.

Algorithms↗

PubMed2005

Effects of tributyltin, benzo[a]pyrene, and their mixture on antioxidant defense systems in Sebastiscus marmoratus.

It has been reported that there is a metabolic interaction between tributyltin (TBT), an organometal used as an antifouling biocide, and benzo[a]pyrene (BaP), a widespread carcinogenic polycyclic aromatic hydrocarbon. This study was therefore designed to examine the potential in vivo influence of TBT, BaP, and their mixture on hepatic antioxidant defense systems of Sebastiscus marmoratus, which were given a single ip injection of TBT (0.5, 1, 5, and 10mg/kg), BaP (0.5, 1, 5, and 10mg/kg), or both in combination (0.5, 1, 5, and 10mg/kg); control fish received olive oil vehicle only. Samples were collected for biochemical analysis after injection for 7 days. Cotreatment with BaP caused a significant inhibition of TBT-mediated malondialdehyde contents elevation. Cotreatment with TBT decreased BaP-mediated glutathione peroxidase activity induction. Cotreatment with TBT and BaP did not significantly alter the reduced glutathione levels, which were significantly induced by TBT or BaP alone. TBT-induced suppression of BaP bioactivation or BaP-induced stimulation of the phase II metabolism of TBT and its biliary excretion, both of which have been reported previously, could explain the observed antagonism. The results suggest that combined exposure of TBT and BaP increases the vulnerability of the fish to oxidative stress. BaP cotreatment decreased the induction of glutathione S-transferase (GST) activity by the lower dose of TBT, while cotreatment with TBT and BaP at the highest dose (10mg/kg) resulted in inhibition of the GST activity by 4.8-fold. The results suggest that these biomarkers should be interpreted with caution in biomonitoring studies. Combined effects of TBT and BaP exposure at environmental levels on these biomarkers should be further researched.

Animals↗

PubMed2005

Tachyplesin activates the classic complement pathway to kill tumor cells.

Tachyplesin is a small, cationic peptide that possesses antitumor properties. However, little is known about its action mechanism. We used phage display to identify a protein that interacted with tachyplesin and isolated a sequence corresponding to the collagen-like domain of C1q, a key component in the complement pathway. Their interaction was subsequently confirmed by both ELISA and affinity precipitation. Tachyplesin seemed to activate the classic complement cascade because it triggered several downstream events, including the cleavage and deposition of C4 and C3 and the formation of C5b-9. When TSU tumor cells were treated with tachyplesin in the presence of serum, activated C4b and C3b could be detected on tumor cells by flow cytometry, Western blotting, and confocal microscopy. However, this effect was blocked when the tumor cells were treated with hyaluronidase or a large excess of hyaluronan, indicating that hyaluronan or related glycosaminoglycans were involved in this process. Treatment of cells with tachyplesin and serum increased in membrane permeability as indicated by the ability of FITC-dextran to enter the cytoplasm. Finally, the combination of tachyplesin and human serum markedly inhibited the proliferation and caused death of TSU cells, and these effects were attenuated if the serum was heat-inactivated or if hyaluronidase was added. Taken together, these observations suggest that tachyplesin binds to both hyaluronan on the cell surface and C1q in the serum and activates the classic complement cascade, which damages the integrity of the membranes of the tumor cells resulting in their death.

Amino Acid Sequence↗

PubMed2005

CLUE: cluster-based retrieval of images by unsupervised learning.

In a typical content-based image retrieval (CBIR) system, target images (images in the database) are sorted by feature similarities with respect to the query. Similarities among target images are usually ignored. This paper introduces a new technique, cluster-based retrieval of images by unsupervised learning (CLUE), for improving user interaction with image retrieval systems by fully exploiting the similarity information. CLUE retrieves image clusters by applying a graph-theoretic clustering algorithm to a collection of images in the vicinity of the query. Clustering in CLUE is dynamic. In particular, clusters formed depend on which images are retrieved in response to the query. CLUE can be combined with any real-valued symmetric similarity measure (metric or nonmetric). Thus, it may be embedded in many current CBIR systems, including relevance feedback systems. The performance of an experimental image retrieval system using CLUE is evaluated on a database of around 60,000 images from COREL. Empirical results demonstrate improved performance compared with a CBIR system using the same image similarity measure. In addition, results on images returned by Google's Image Search reveal the potential of applying CLUE to real-world image data and integrating CLUE as a part of the interface for keyword-based image retrieval systems.

Algorithms↗

PubMed2004

Molecular interaction of retinoic acid receptors with coregulators PCAF and RIP140.

p300/CBP-associating factor (PCAF) is a ligand-dependent coactivator, whereas receptor-interacting protein 140 (RIP140) is a ligand-dependent negative coregulator for retinoic acid (RA) receptor (RAR) and retinoid X receptor (RXR). To compare these molecular interactions and to determine the effect of RXR ligands, we focus on PCAF/RAR/RXR complex formation in this study for a comparison to RIP140/RAR/RXR complex formation. The LBD of RXR is identified as its primary PCAF-interacting motif. BIAcore studies determine the Kd of RAR/RXR association with PCAF as 9.35 nM in the presence of RXR ligand AGN194204, and 47.2 nM in the absence of ligand. Cross-linking study demonstrates tri-molecular complex consisting of one RAR/RXR pair and one PCAF. In competition experiments, RIP140 strongly competes with PCAF for interaction with RAR/RXR both in vitro and in vivo. Chromatin immunoprecipitation demonstrates recruitment of RIP140 and PCAF to the endogenous RA-regulated gene, the RARbeta2 promoter. This study presents kinetic evidence for competition of RIP140 with PCAF for ligand-dependent interactions with RAR/RXR, and provides kinetic explanation for the suppressive activity of RIP140 in RA-activated gene expression.

Acetyltransferases↗

PubMed2004

Genome sequence of the Brown Norway rat yields insights into mammalian evolution.

The laboratory rat (Rattus norvegicus) is an indispensable tool in experimental medicine and drug development, having made inestimable contributions to human health. We report here the genome sequence of the Brown Norway (BN) rat strain. The sequence represents a high-quality 'draft' covering over 90% of the genome. The BN rat sequence is the third complete mammalian genome to be deciphered, and three-way comparisons with the human and mouse genomes resolve details of mammalian evolution. This first comprehensive analysis includes genes and proteins and their relation to human disease, repeated sequences, comparative genome-wide studies of mammalian orthologous chromosomal regions and rearrangement breakpoints, reconstruction of ancestral karyotypes and the events leading to existing species, rates of variation, and lineage-specific and lineage-independent evolutionary events such as expansion of gene families, orthology relations and protein evolution.

Animals↗

PubMed2003

Suppressive effect of receptor-interacting protein 140 on coregulator binding to retinoic acid receptor complexes, histone-modifying enzyme activity, and gene activation.

Gene induction by retinoic acid (RA) is suppressed by overexpression of receptor-interacting protein 140 (RIP140). RIP140-mediated suppression was reversed most effectively by overexpressing the coactivator p300/CREB-binding protein-associated factor (P/CAF). Immunoprecipitation demonstrated coexistence of holoreceptors complexed with RIP140 or P/CAF. Chromatin immunoprecipitation revealed rapid RA-enhanced recruitment of RIP140, but delayed P/CAF recruitment, to an RA-targeted promoter in COS-1 cells supplemented with RIP140. In RA-induced P19 cells, endogenous RIP140 was rapidly (within 4 h) and significantly recruited to both the RARbeta2 and TR2 genes, whereas the peak of endogenous P/CAF recruitment occurred much later (48 h) and to a lesser degree. Consistent with these observations, significant histone acetylation of endogenous RA receptor (RAR) targets was only observed 48 h following RA treatment. In vitro experiments confirmed RA-induced transcription from a chromatin template, which was reduced by adding RIP140. This study presents evidence for coexistence of multiple RAR-coregulator complexes and a preferential RA-induced recruitment of RIP140 to endogenous RAR-targeted promoters after short term RA treatment, which correlates with suppressed induction of RA-regulated gene expression in the presence of RIP140.

Acetyltransferases↗

PubMed2003

A peptide with three hyaluronan binding motifs inhibits tumor growth and induces apoptosis.

A number of hyaluronan (HA) binding proteins such as soluble CD44, receptor for hyaluronan-mediated motility (RHAMM), and metastatin inhibit tumor growth and metastasis. To determine whether the HA binding motif is the element responsible for the antitumor effect of this family of proteins, we examined the biological activity of a 42-amino acid peptide (designated as BH-P) that contains three HA binding motifs [B(X(7))B] from human brain HA binding protein. In initial experiments with cultured cells, we found that synthetic BH-P inhibited the proliferation and colony formation of tumor cells. It also blocked the growth of tumors on the chorioallantoic membranes of 10-day chicken embryos. In addition, MDA-435 melanoma cells that had been transfected with an expression vector for BH-P grew at a slower rate in nude mice than the vector-alone transfectants. Final studies revealed that the BH-P could activate caspase-8, caspase-3, and poly(ADP-ribose) polymerase and trigger the apoptosis of tumor cells. Taken together, these results suggest that the HA binding motif that is present in HA binding proteins may be responsible for the antitumor effect exerted by the members of this family.

Amino Acid Motifs↗

PubMed2003

Application of artificial neural networks in the design of controlled release drug delivery systems.

Controlled release drug delivery systems offer great advantages over the conventional dosage forms. However, there are great challenges to efficiently develop controlled release drug delivery systems due to the complexity of these delivery systems. Traditional statistic response surface methodology (RSM) is one of the techniques that has been employed to develop and formulate controlled release dosage forms. However, there are some limitations to the RSM technique. Hence, another technique called artificial neural networks (ANN) has recently gained wide popularity in the development of controlled release dosage forms. In this review, the basic ANN structure, the development of the ANN model and an explanation of how to use ANN to design and develop controlled release drug delivery systems are discussed. In addition, the applications of ANN in the design and development of controlled release dosage forms are also summarized in this review.

Delayed-Action Preparations↗

PubMed2002

Comparison of four artificial neural network software programs used to predict the in vitro dissolution of controlled-release tablets.

The purpose of this study was to evaluate four commercially available artificial neural network (ANN) software programs: NeuroShell2 v3.0, BrainMaker v3.7, CAD/Chem v5.0, and NeuralWorks Professional II/Plus for prediction of in vitro dissolution-time profiles of controlled-release tablets containing a model sympathomimetic drug. Seven independent formulation variables and three other tablet variables (moisture content of granules, granule particle size, and tablet hardness), for 22 tablet formulations, were used as the ANN model input. In vitro dissolution time-profiles at 10 different sampling times were used as the output. The models' optimum architectures were determined for each ANN software by varying the number of hidden layers and number of nodes in hidden layer(s). The ANN developed from the four software programs were validated by predicting the in vitro dissolution time-profiles of each of the 19 formulations, which were excluded from the training process. Although the same data set was used, the optimum ANN architectures generated from the four software programs were different. Using the four optimum ANN models, the plots of predicted vs. observed percentage of drug dissolved gave slopes ranging from 0.95 to 1.01 and r2 values ranging from 0.95 to 0.99 for all 190 dissolution data points for the 19 training formulations. The difference factors (f1) and similarity factors (f2) between the ANN predicted and the observed in vitro dissolution profiles were also used to compare the predictions for the four software programs. It was concluded that the four programs provided reasonable predictions of in vitro dissolution profiles for the data set employed in this study, with NeuralShell2 showing the best overall prediction.

Delayed-Action Preparations↗

PubMed2002

Real-time analysis of molecular interaction of retinoid receptors and receptor-interacting protein 140 (RIP140).

Receptor interacting protein 140 (RIP140) is a coregulator for a large number of transcription factors. RIP140 interacts with retinoic acid receptor (RAR) and retinoid X receptor (RXR) with or without ligands. The C-terminal domain of RIP140 (RIP-C') contains a novel sequence (1063-1076, LTKTNPILYYMLQK) and has been shown to interact with RAR and RXR ligand dependently in two-hybrid interaction and pull-down assays. To examine the kinetic characteristics of molecular interaction of RIP-C' with RAR and RXR, a surface plasmon resonance technology (BIAcore) was applied for real-time analyses of this molecular interaction with highly purified proteins. A modified pull-down assay using purified proteins was also conducted to obtain supporting data. The effect of retinoid ligands on this type of interaction was addressed. By using receptor mutants, it was demonstrated that the activation function-2 domain and the ability to form dimers of the receptors are required for an efficient interaction of receptor with RIP140. Finally, with a mutagenesis approach, we determined the effects of specific point mutations on the kinetics of RIP-C' interaction with RAR/RXR.

Adaptor Proteins, Signal Transducing↗