Binding peptide generation for MHC Class I proteins with deep reinforcement learning.

MOTIVATION: MHC Class I protein plays an important role in immunotherapy by presenting immunogenic peptides to anti-tumor immune cells. The repertoires of peptides for various MHC Class I proteins are distinct, which can be reflected by their diverse binding motifs. To characterize binding motifs for MHC Class I proteins, in vitro experiments have been conducted to screen peptides with high binding affinities to hundreds of given MHC Class I proteins. However, considering tens of thousands of known MHC Class I proteins, conducting in vitro experiments for extensive MHC proteins is infeasible, and thus a more efficient and scalable way to characterize binding motifs is needed. RESULTS: We presented a de novo generation framework, coined PepPPO, to characterize binding motif for any given MHC Class I proteins via generating repertoires of peptides presented by them. PepPPO leverages a reinforcement learning agent with a mutation policy to mutate random input peptides into positive presented ones. Using PepPPO, we characterized binding motifs for around 10 000 known human MHC Class I proteins with and without experimental data. These computed motifs demonstrated high similarities with those derived from experimental data. In addition, we found that the motifs could be used for the rapid screening of neoantigens at a much lower time cost than previous deep-learning methods. AVAILABILITY AND IMPLEMENTATION: The software can be found in https://github.com/minrq/pMHC. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Detection of Cognitive Impairment From eSAGE Metadata Using Machine Learning.

OBJECTIVE: Using the metadata collected in the digital version of the Self-Administered Gerocognitive Examination (eSAGE), we aim to improve the prediction of mild cognitive impairment (MCI) and dementia (DM) by applying machine learning methods. PATIENTS AND METHODS: A total of 66 patients had a diagnosis of normal cognition (NC), MCI, or DM, and eSAGE scores and metadata were used. eSAGE scores and metadata were obtained. Each eSAGE question was scored and behavioral features (metadata) such as the time spent on each test page, drawing speed, and average stroke length were extracted for each patient. Logistic regression (LR) and gradient boosting models were trained using these features to detect cognitive impairment (CI). Performance was evaluated using 10-fold cross-validation, with accuracy, precision, recall, F1 score, and receiver operating characteristic area under the curve (AUC) score as evaluation metrics. RESULTS: LR with feature selection achieved an AUC of 89.51%, a recall of 87.56%, and an F1 of 85.07% using both behavioral and scoring. LR using scores and metadata also achieved an AUC of 84.00% in detecting MCI from NC, and an AUC of 98.12% in detecting DM from NC. Average stroke length was particularly useful for prediction and when combined with 4 other scoring features, LR achieved an even better AUC of 92.06% in detecting CI. The study shows that eSAGE scores and metadata are predictive of CI. CONCLUSIONS: eSAGE scores and metadata are predictive of CI. With machine learning methods, the metadata could be combined with scores to enable more accurate detection of CI.

Improving Anticancer Drug Selection and Prioritization via Neural Learning to Rank.

Personalized cancer treatment requires a thorough understanding of complex interactions between drugs and cancer cell lines in varying genetic and molecular contexts. To address this, high-throughput screening has been used to generate large-scale drug response data, facilitating data-driven computational models. Such models can capture complex drug-cell line interactions across various contexts in a fully data-driven manner. However, accurately prioritizing the most effective drugs for each cell line still remains a significant challenge. To address this, we developed multiple neural ranking approaches that leverage large-scale drug response data across multiple cell lines from diverse cancer types. Unlike existing approaches that primarily utilize regression and classification techniques for drug response prediction, we formulated the objective of drug selection and prioritization as a drug ranking problem. In this work, we proposed multiple pairwise and listwise neural ranking methods that learn latent representations of drugs and cell lines and then use those representations to score drugs in each cell line via a learnable scoring function. Specifically, we developed neural pairwise and listwise ranking methods, Pair-PushC and List-One on top of the existing methods, pLETORg and ListNet, respectively. Additionally, we proposed a novel listwise ranking method, List-All, that focuses on all the effective drugs instead of the top effective drug, unlike List-One. We also provide an exhaustive empirical evaluation with state-of-the-art regression and ranking baselines on large-scale data sets across multiple experimental settings. Our results demonstrate that our proposed ranking methods mostly outperform the best baselines with significant improvements of as much as 25.6% in terms of selecting truly effective drugs within the top 20 predicted drugs (i.e., hit@20) across 50% test cell lines. Furthermore, our analyses suggest that the learned latent spaces from our proposed methods demonstrate informative clustering structures and capture relevant underlying biological features. Moreover, our comprehensive evaluation provides a thorough and objective comparison of the performance of different methods (including our proposed ones).

RLSynC: Offline-Online Reinforcement Learning for Synthon Completion.

Retrosynthesis is the process of determining the set of reactant molecules that can react to form a desired product. Semitemplate-based retrosynthesis methods, which imitate the reverse logic of synthesis reactions, first predict the reaction centers in the products and then complete the resulting synthons back into reactants. We develop a new offline-online reinforcement learning method RLSynC for synthon completion in semitemplate-based methods. RLSynC assigns one agent to each synthon, all of which complete the synthons by conducting actions step by step in a synchronized fashion. RLSynC learns the policy from both offline training episodes and online interactions, which allows RLSynC to explore new reaction spaces. RLSynC uses a standalone forward synthesis model to evaluate the likelihood of the predicted reactants in synthesizing a product and thus guides the action search. Our results demonstrate that RLSynC can outperform state-of-the-art synthon completion methods with improvements as high as 14.9%, highlighting its potential in synthesis planning.

Retinal Degeneration Response to Graphene Quantum Dots: Disruption of the Blood-Retina Barrier Modulated by Surface Modification-Dependent DNA Methylation.

Graphene quantum dots (GQDs) are used in diverse fields from chemistry-related materials to biomedicines, thus causing their substantial release into the environment. Appropriate visual function is crucial for facilitating the decision-making process within the nervous system. Given the direct interaction of eyes with the environment and even nanoparticles, herein, GQDs, sulfonic acid-doped GQDs (S-GQDs), and amino-functionalized GQDs (A-GQDs) were employed to understand the potential optic neurotoxicity disruption mechanism by GQDs. The negatively charged GQDs and S-GQDs disturbed the response to light stimulation and impaired the structure of the retinal nuclear layer of zebrafish larvae, causing vision disorder and retinal degeneration. Albeit with sublethal concentrations, a considerably reduced expression of the retinal vascular sprouting factor sirt1 through increased DNA methylation damaged the blood-retina barrier. Importantly, the regulatory effect on vision function was influenced by negatively charged GQDs and S-GQDs but not positively charged A-GQDs. Moreover, cluster analysis and computational simulation studies indicated that binding affinities between GQDs and the DNMT1-ligand binding might be the dominant determinant of the vision function response. The previously unknown pathway of blood-retinal barrier interference offers opportunities to investigate the biological consequences of GQD-based nanomaterials, guiding innovation in the industry toward environmental sustainability.

Using Machine Learning and Multifaceted Preoperative Measures to Predict Adult Cochlear Implant Outcomes: A Prospective Pilot Study.

OBJECTIVES: To use machine learning and a battery of measures for preoperative prediction of speech recognition and quality of life (QOL) outcomes after cochlear implant (CI) surgery. DESIGN: Demographic, audiologic, cognitive-linguistic, and QOL predictors were collected from 30 postlingually deaf adults before CI surgery. K-means clustering separated patients into groups. Reliable change index scores were computed for speech recognition and QOL from pre-CI to 6 months post-CI, and group differences were determined. RESULTS: Clustering yielded three groups with differences in reliable change index for sentence recognition. One group demonstrated low baseline sentence recognition and only small improvements post-CI, suggesting a group "at risk" for limited benefits. This group showed lower pre-CI scores on verbal learning and memory and lack of musical training. CONCLUSIONS: Preoperative assessments can prognosticate CI recipients' postoperative performance and identify individuals at risk for experiencing poor sentence recognition outcomes, which may help guide counseling and rehabilitation.

Triazole fungicides disrupt embryonic stem cell differentiation: Potential modulatory role of the retinoic acid signaling pathway.

The developmental toxicity and human health risks of triazole fungicides (TFs) have attracted worldwide attention due to the ability to enter the human body in a variety of ways. Nevertheless, the specific mechanism by which TFs exert remains incompletely understood. Given that retinoic acid (RA) signaling pathway are closely related to development, this study aimed to screen and identify developmentally disabled chemicals in commonly used TFs and to reveal the potential effects of TFs on developmental retardation through the RA signaling pathway in mouse embryonic stem cells (mESCs). Specifically, six typical TFs (myclobutanil, tebuconazole, hexaconazole, propiconazole, difenoconazole, and flusilazole) were exposed through the construction of an embryoid bodies (EBs)-based in vitro global differentiation models. Our results clarified that various TFs disturbed lineage commitment during early embryonic development. Crucially, the activation of RA signaling pathway, which alters the expression of key genes and interferes the transport and metabolism of retinol, may be responsible for this effect. Furthermore, molecular docking, molecular dynamics simulations, and experiments using a retinoic acid receptor α inhibitor provide evidence supporting the potential modulatory role of the retinoic acid signaling pathway in developmental injury. The current study offers new insights into the TFs involved in the RA signaling pathway that interfere with the differentiation process of mESCs, which is crucial for understanding the impact of TFs on pregnancy and early development.

Tumor necrosis factor α1 decreases mucosal immune and antioxidant response in the midgut of hybrid fish (white crucian carp ♀ × red crucian carp ♂).

Tumor necrosis factor α1 (TNFα) is a pleiotropic cytokine involved in immune regulation and cellular homeostasis, but the crucial role of TNFα in fish gut remained unclear. The current study aimed to evaluate the immunoregulatory function of TNFα1 on gut barrier in a novel hybrid fish (WR), which was produced by crossing white crucian carp (Carassius cuvieri, ♀) with red crucian carp (Carassius auratus red var, ♂). In this study, WR-tnfα1 sequence was identified, and a high-level expression was detected in the intestine. Elevated levels of WR-tnfα1 expressions were detected in immune-related tissues and cultured fish cells on stimulation. The appearance of vacuolization and submucosal rupture was observed in TNFα1-treated midgut of WR, along with elevated levels of goblet cell atrophy, whereas no significant changes were detected in most expressions of tight-junction genes and mucin genes. In contrast, WR receiving gut perfusion with WR-TNFα1 showed a remarkable decrease in antioxidant status in midgut, whereas the expression levels of apoptotic genes and redox responsive genes increased sharply. These results suggested that TNFα1 could exhibit a detrimental effect on antioxidant defense and immune regulation in the midgut of WR.

Development of complemented comprehensive networks for rapid screening of repurposable drugs applicable to new emerging disease outbreaks.

BACKGROUND: Computational drug repurposing is crucial for identifying candidate therapeutic medications to address the urgent need for developing treatments for newly emerging infectious diseases. The recent COVID-19 pandemic has taught us the importance of rapidly discovering candidate drugs and providing them to medical and pharmaceutical experts for further investigation. Network-based approaches can provide repurposable drugs quickly by leveraging comprehensive relationships among biological components. However, in a case of newly emerging disease, applying a repurposing methods with only pre-existing knowledge networks may prove inadequate due to the insufficiency of information flow caused by the novel nature of the disease. METHODS: We proposed a network-based complementary linkage method for drug repurposing to solve the lack of incoming new disease-specific information in knowledge networks. We simulate our method under the controlled repurposing scenario that we faced in the early stage of the COVID-19 pandemic. First, the disease-gene-drug multi-layered network was constructed as the backbone network by fusing comprehensive knowledge database. Then, complementary information for COVID-19, containing data on 18 comorbid diseases and 17 relevant proteins, was collected from publications or preprint servers as of May 2020. We estimated connections between the novel COVID-19 node and the backbone network to construct a complemented network. Network-based drug scoring for COVID-19 was performed by applying graph-based semi-supervised learning, and the resulting scores were used to validate prioritized drugs for population-scale electronic health records-based medication analyses. RESULTS: The backbone networks consisted of 591 diseases, 26,681 proteins, and 2,173 drug nodes based on pre-pandemic knowledge. After incorporating the 35 entities comprised of complemented information into the backbone network, drug scoring screened top 30 potential repurposable drugs for COVID-19. The prioritized drugs were subsequently analyzed in electronic health records obtained from patients in the Penn Medicine COVID-19 Registry as of October 2021 and 8 of these were found to be statistically associated with a COVID-19 phenotype. CONCLUSION: We found that 8 of the 30 drugs identified by graph-based scoring on complemented networks as potential candidates for COVID-19 repurposing were additionally supported by real-world patient data in follow-up analyses. These results show that our network-based complementary linkage method and drug scoring algorithm are promising strategies for identifying candidate repurposable drugs when new emerging disease outbreaks.

Alteration of synergistic immune response in gut-liver axis of white crucian carp (Carassius cuvieri) after gut infection with Aeromonas hydrophila.

Aeromonas hydrophila can pose a great threat to the survival of farmed fish. In current study, we investigated the pathological characteristics and immune response in gut-liver axis of white crucian carp (WCC) upon gut infection. WCC anally intubated with A. hydrophila exerted a tissue deformation in damaged midgut with elevated levels of goblet cells along with a significant decrease in tight junction proteins and villi length-to-width ratios. In addition, immune-related gene expressions and antioxidant properties increased dramatically in gut-liver axis of WCC following gut infection with A. hydrophila. These results highlighted the immune modulation and redox alteration in gut-liver axis of WCC in response to gut infection.

M2: Mixed Models With Preferences, Popularities and Transitions for Next-Basket Recommendation.

Next-basket recommendation considers the problem of recommending a set of items into the next basket that users will purchase as a whole. In this paper, we develop a novel mixed model with preferences, popularities and transitions (M2) for the next-basket recommendation. This method models three important factors in next-basket generation process: 1) users' general preferences, 2) items' global popularities and 3) transition patterns among items. Unlike existing recurrent neural network-based approaches, M2 does not use the complicated networks to model the transitions among items, or generate embeddings for users. Instead, it has a simple encoder-decoder based approach (ed-Trans) to better model the transition patterns among items. We compared M2 with different combinations of the factors with 5 state-of-the-art next-basket recommendation methods on 4 public benchmark datasets in recommending the first, second and third next basket. Our experimental results demonstrate that M2 significantly outperforms the state-of-the-art methods on all the datasets in all the tasks, with an improvement of up to 22.1%. In addition, our ablation study demonstrates that the ed-Trans is more effective than recurrent neural networks in terms of the recommendation performance. We also have a thorough discussion on various experimental protocols and evaluation metrics for next-basket recommendation evaluation.

[Network Meta-analysis of oral Chinese patent medicines in treatment of acute exacerbation of chronic obstructive pulmonary disease].

This study aimed to evaluate the effectiveness and safety of eight oral Chinese patent medicines in the treatment of acute exacerbation of chronic obstructive pulmonary disease(AECOPD) by network Meta-analysis. Randomized controlled trial(RCT) on the treatment of AECOPD with eight oral Chinese patent medicines was retrieved from databases including CNKI, Wanfang, VIP, SinoMed, PubMed, Web of Science, EMbase, and Cochrane Library from database inception to August 6, 2022. The information was extracted from the included literature and the quality of the included studies was evaluated using the Cochrane risk of bias assessment tool. The data were analyzed using Stata SE 15.1 and ADDIS 1.16.8 software. Finally, 53 RCTs were included, with 5 289 patients involved, including 2 652 patients in the experimental group and 2 637 patients in the control group. Network Meta-analysis showed that Lianhua Qingwen Capsules+conventional western medicine were optimal in improving clinical effective rate, Shufeng Jiedu Capsules+conventional western medicine in improving FEV1/FVC, Qingqi Huatan Pills+conventional western medicine in improving FEV1%pred, Feilike Mixture(Capsules)+conventional western medicine in improving PaO_2, Lianhua Qingwen Capsules+conventional western medicine in reducing PaCO_2, and Qingqi Huatan Pills+conventional western medicine in reducing C-reactive protein(CRP). In terms of safety, most of them were gastrointestinal symptoms, and no serious adverse reactions were reported. When the clinical effective rate was taken as the comprehensive index of efficacy evaluation, Lianhua Qingwen Capsules+conventional western medicine were the most likely to be the best treatment for AECOPD. There are some limitations in the conclusion of this study. It only provides references for clinical medication.

Evaluation of Machine Learning Models for Proteoform Retention and Migration Time Prediction in Top-Down Mass Spectrometry.

Reversed-phase liquid chromatography (RPLC) and capillary zone electrophoresis (CZE) are two primary proteoform separation methods in mass spectrometry (MS)-based top-down proteomics. Proteoform retention time (RT) prediction in RPLC and migration time (MT) prediction in CZE provide additional information for accurate proteoform identification and quantification. While existing methods are mainly focused on peptide RT and MT prediction in bottom-up MS, there is still a lack of methods for proteoform RT and MT prediction in top-down MS. We systematically evaluated eight machine learning models and a transfer learning method for proteoform RT prediction and five models and the transfer learning method for proteoform MT prediction. Experimental results showed that a gated recurrent unit (GRU)-based model with transfer learning achieved a high accuracy (R = 0.978) for proteoform RT prediction and that the GRU-based model and a fully connected neural network model obtained a high accuracy of R = 0.982 and 0.981 for proteoform MT prediction, respectively.

A pharmacovigilance study of pharmacokinetic drug interactions using a translational informatics discovery approach.

BACKGROUND: While the pharmacokinetic (PK) mechanisms for many drug interactions (DDIs) have been established, pharmacovigilance studies related to these PK DDIs are limited. Using a large surveillance database, a translational informatics approach can systematically screen adverse drug events (ADEs) for many DDIs with known PK mechanisms. METHODS: We collected a set of substrates and inhibitors related to the cytochrome P450 (CYP) isoforms, as recommended by the United States Food and Drug Administration (FDA) and Drug Interactions Flockhart table™. The FDA's Adverse Events Reporting System (FAERS) was used to obtain ADE reports from 2004 to 2018. The substrate and inhibitor information were used to form PK DDI pairs for each of the CYP isoforms and Medical Dictionary for Regulatory Activities (MedDRA) preferred terms used for ADEs in FAERS. A shrinkage observed-to-expected ratio (Ω) analysis was performed to screen for potential PK DDI and ADE associations. RESULTS: We identified 149 CYP substrates and 62 CYP inhibitors from the FDA and Flockhart tables. Using FAERS data, only those DDI-ADE associations were considered that met the disproportionality threshold of Ω > 0 for a CYP substrate when paired with at least two inhibitors. In total, 590 ADEs were associated with 2085 PK DDI pairs and 38 individual substrates, with ADEs overlapping across different CYP substrates. More importantly, we were able to find clinical and experimental evidence for the paclitaxel-clopidogrel interaction associated with peripheral neuropathy in our study. CONCLUSION: In this study, we utilized a translational informatics approach to discover potentially novel CYP-related substrate-inhibitor and ADE associations using FAERS. Future clinical, population-based and experimental studies are needed to confirm our findings.

Comparative analysis on the immunoregulatory roles of ferritin M in hybrid fish (Carassius cuvieri ♀ × Carassius auratus red var ♂) and its parental species after bacterial infection.

Ferritin M is involved in the regulation of fish immunity. In this study, open reading frame (ORF) sequences of ferritin M from hybrid fish and its parental species were 534 bp. Tissue-specific analysis indicated that the highest level of ferritin M from red crucian carp was observed in kidney, while peaked expressions of ferritin M from white crucian carp and hybrid carp were observed in gill. Elevated levels of ferritin M from hybrid carp and its parental species were detected in immune-related tissues following Aeromonas hydrophila infection or in cultured fish cell lines after lipopolysaccharide (LPS) challenge. Ferritin M overexpression could attenuate NF-κB and TNFα promoter activity in their respective fish cells. Purified ferritin M fusion proteins elicited in vitro binding activity to A. hydrophila and Edwardsiella tarda, lowered bacterial dissemination to tissues and alleviated inflammatory response. Furthermore, treatment with ferritin M fusion proteins could mitigate bacteria-induced liver damage and rescue antioxidant activity. These results suggested that ferritin M in hybrid fish showed a similar immune defense against bacteria infection in comparison with those of its parental species.

Blood cell characterization and transcriptome analysis reveal distinct immune response and host resistance of different ploidy cyprinid fish following Aeromonas hydrophila infection.

Aeromonas hydrophila can pose a great threat to survival of freshwater fish. In this study, A. hydrophila infection could decrease blood cell numbers, promote blood cell damage as well as alter the levels of alkaline phosphatase (ALP), lysozyme (LZM), aspartate aminotransferase (AST), total antioxidant capacity (T-AOC), total superoxide dismutase (SOD), catalase (CAT) and malondialdehyde (MDA) in immune-related tissues of red crucian carp (RCC, 2 N = 100) and triploid cyprinid fish (3 N fish, 3 N = 150). In addition, the significant alternation of antioxidant status was observed in PBMCs isolated from RCC and 3 N following LPS stimulation. The core differential expression genes (DEGs) involved in apoptosis, immunity, inflammation and cellular signals were co-expressed differentially in RCC and 3 N following A. hydrophila challenge. NOD-like receptor (NLR) signals appeared to play a critical role in A. hydrophila-infected fish. DEGs of NLR signals in RCCah vs RCCctl were enriched in caspase-1-dependent Interleukin-1ß (IL-1ß) secretion, interferon (IFN) signals as well as cytokine activation, while DEGs of NLR signals in 3Nah vs 3Nctl were enriched in caspase-1-dependent IL-1ß secretion and antibacterial autophagy. These results highlighted the differential signal regulation of different ploidy cyprinid fish to cope with bacterial infection.

A novel ferritin L (FerL) in hybrid crucian carp could participate in host defense against Aeromonas hydrophila infection and diminish inflammatory signals.

FerL, a multifunctional iron-storage polypeptide, not only exhibited a regulatory role in iron metabolism, but also participated in the regulation of fish immunity. In this study, ORF sequence of WR-FerL was 522 bp, encoding 173 amino acid residues. Tissue-specific analysis revealed that the highest expression of WR-FerL was detected in spleen. A. hydrophila challenge and LPS stimulation could sharply enhance WR-FerL mRNA expression in tissues and fish cells, respectively. Purified WR-FerL fusion peptide exhibited in vitro binding activity to A. hydrophila and endotoxin, limited bacterial dissemination to tissues as well as attenuated A. hydrophila-induced production of pro-inflammatory cytokines. Moreover, WR-FerL overexpression could abrogate NF-κB and TNFα promoter activity in fish cells. These results indicated that WR-FerL could play an important role in host defense against A. hydrophila infection.

Predicting pharmacotherapeutic outcomes for type 2 diabetes: An evaluation of three approaches to leveraging electronic health record data from multiple sources.

Electronic health record (EHR) data are increasingly used to develop prediction models to support clinical care, including the care of patients with common chronic conditions. A key challenge for individual healthcare systems in developing such models is that they may not be able to achieve the desired degree of robustness using only their own data. A potential solution-combining data from multiple sources-faces barriers such as the need for data normalization and concerns about sharing patient information across institutions. To address these challenges, we evaluated three alternative approaches to using EHR data from multiple healthcare systems in predicting the outcome of pharmacotherapy for type 2 diabetes mellitus(T2DM). Two of the three approaches, named Selecting Better (SB) and Weighted Average(WA), allowed the data to remain within institutional boundaries by using pre-built prediction models; the third, named Combining Data (CD), aggregated raw patient data into a single dataset. The prediction performance and prediction coverage of the resulting models were compared to single-institution models to help judge the relative value of adding external data and to determine the best method to generate optimal models for clinical decision support. The results showed that models using WA and CD achieved higher prediction performance than single-institution models for common treatment patterns. CD outperformed the other two approaches in prediction coverage, which we defined as the number of treatment patterns predicted with an Area Under Curve of 0.70 or more. We concluded that 1) WA is an effective option for improving prediction performance for common treatment patterns when data cannot be shared across institutional boundaries and 2) CD is the most effective approach when such sharing is possible, especially for increasing the range of treatment patterns that can be predicted to support clinical decision making.

Transcriptome analysis and co-expression network reveal the mechanism linking mitochondrial function to immune regulation in red crucian carp (Carassius auratus red var) after Aeromonas hydrophila challenge.

Aeromonas hydrophila is a common pathogen of freshwater fish. In this study, A. hydrophila infection was shown to cause tissue damage, trigger physiological changes as well as alter the expression profiles of immune- and metabolic-related genes in immune tissues of red crucian carp (RCC). Transcriptome analysis revealed that acute A. hydrophila infection exerted a profound effect on mitochondrial oxidative phosphorylation linking metabolic regulation to immune response. In addition, we further identified cellular senescence, apoptosis, necrosis and mitogen-activated protein kinase signal pathways as crucial signal pathways in the kidney of RCC subjected to A. hydrophila infection. These findings may have important implications for understanding modulation of immunometabolic response to bacterial infection.

HAM: Hybrid Associations Models for Sequential Recommendation.

Sequential recommendation aims to identify and recommend the next few items for a user that the user is most likely to purchase/review, given the user's purchase/rating trajectories. It becomes an effective tool to help users select favorite items from a variety of options. In this manuscript, we developed hybrid associations models (HAM) to generate sequential recommendations. using three factors: 1) users' long-term preferences, 2) sequential, high-order and low-order association patterns in the users' most recent purchases/ratings, and 3) synergies among those items. HAM uses simplistic pooling to represent a set of items in the associations, and element-wise product to represent item synergies of arbitrary orders. We compared HAM models with the most recent, state-of-the-art methods on six public benchmark datasets in three different experimental settings. Our experimental results demonstrate that HAM models significantly outperform the state of the art in all the experimental settings. with an improvement as much as 46.6%. In addition, our run-time performance comparison in testing demonstrates that HAM models are much more efficient than the state-of-the-art methods. and are able to achieve significant speedup as much as 139.7 folds.