Sequential Pattern Research Articles

DeepHLAPred: a deep learning-based method for non-classical HLA binder prediction

Human leukocyte antigen (HLA) is closely involved in regulating the human immune system. Despite great advance in detecting classical HLA Class I binders, there are few methods or toolkits for recognizing non-classical HLA Class I binders. To fill in this gap, we have developed a deep learning-based tool called DeepHLAPred. The DeepHLAPred used electron-ion interaction pseudo potential, integer numerical mapping and accumulated amino acid frequency as initial representation of non-classical HLA binder sequence. The deep learning module was used to further refine high-level representations. The deep learning module comprised two parallel convolutional neural networks, each followed by maximum pooling layer, dropout layer, and bi-directional long short-term memory network. The experimental results showed that the DeepHLAPred reached the state-of-the-art performanceson the cross-validation test and the independent test. The extensive test demonstrated the rationality of the DeepHLAPred. We further analyzed sequence pattern of non-classical HLA class I binders by information entropy. The information entropy of non-classical HLA binder sequence implied sequence pattern to a certain extent. In addition, we have developed a user-friendly webserver for convenient use, which is available at http://www.biolscience.cn/DeepHLApred/. The tool and the analysis is helpful to detect non-classical HLA Class I binder. The source code and data is available at https://github.com/tangxingyu0/DeepHLApred.

Polyreactivity of antibodies from different B-cell subpopulations is determined by distinct sequence patterns of variable region.

An antibody molecule that can bind to multiple distinct antigens is defined as polyreactive. In the present study, we performed statistical analyses to assess sequence correlates of polyreactivity of >600 antibodies cloned from different B-cell types of healthy humans. The data revealed several sequence patterns of variable regions of heavy and light immunoglobulin chains that determine polyreactivity. The most prominent identified patterns were increased number of basic amino acid residues, reduced frequency of acidic residues, increased number of aromatic and hydrophobic residues, and longer length of CDR L1. Importantly, our study revealed that antibodies isolated from different B-cell populations used distinct sequence patterns (or combinations of them) for polyreactive antigen binding. Furthermore, we combined the data from sequence analyses with molecular modeling of selected polyreactive antibodies and demonstrated that human antibodies can use multiple pathways for achieving antigen-binding promiscuity. These data reconcile some contradictions in the literature regarding the determinants of antibody polyreactivity. Moreover, our study demonstrates that the mechanism of polyreactivity of antibodies evolves during immune response and might be tailored to specific functional properties of different B-cell compartments. Finally, these data can be of use for efforts in the development and engineering of therapeutic antibodies.

A psychometric evaluation of the NIH Toolbox fluid cognition tests adapted for Swahili and Dholuo languages in Kenyan children and adolescents.

Our objective was to evaluate the psychometric properties of the culturally adapted NIH Toolbox African Languages® when used in Swahili and Dholuo-speaking children in western Kenya. Swahili-speaking participants were recruited from Eldoret and Dholuo-speaking participants from Ajigo; all were <14 years of age and enrolled in primary school. Participants completed a demographics questionnaire and five fluid cognition tests of the NIH Toolbox® African Languages program, including Flanker, Dimensional Change Card Sort (DCCS), Picture Sequence Memory, Pattern Comparison, and List Sorting tests. Statistical analyses examined aspects of reliability, including internal consistency (in both languages) and test-retest reliability (in Dholuo only). Participants included 479 children (n = 239, Swahili-speaking; n = 240, Dholuo-speaking). Generally, the tests had acceptable psychometric properties for research use within Swahili- and Dholuo-speaking populations (mean age = 10.5; SD = 2.3). Issues related to shape identification and accuracy over speed limited the utility of DCCS for many participants, with approximately 25% of children unable to match based on shape. These cultural differences affected outcomes of reliability testing among the Dholuo-speaking cohort, where accuracy improved across all five tests, including speed. There is preliminary evidence that the NIH Toolbox ® African Languages potentially offers a valid assessment of development and performance using tests of fluid cognition in Swahili and Dholuo among research settings. With piloting underway across other diverse settings, future research should gather additional evidence on the clinical utility and acceptability of these tests, specifically through the establishment of norming data among Kenyan regions and evaluating these psychometric properties.

Evidential deep learning for trustworthy prediction of enzyme commission number.

The rapid growth of uncharacterized enzymes and their functional diversity urge accurate and trustworthy computational functional annotation tools. However, current state-of-the-art models lack trustworthiness on the prediction of the multilabel classification problem with thousands of classes. Here, we demonstrate that a novel evidential deep learning model (named ECPICK) makes trustworthy predictions of enzyme commission (EC) numbers with data-driven domain-relevant evidence, which results in significantly enhanced predictive power and the capability to discover potential new motif sites. ECPICK learns complex sequential patterns of amino acids and their hierarchical structures from 20million enzyme data. ECPICK identifies significant amino acids that contribute to the prediction without multiple sequence alignment. Our intensive assessment showed not only outstanding enhancement of predictive performance on the largest databases of Uniprot, Protein Data Bank (PDB) and Kyoto Encyclopedia of Genes and Genomes (KEGG), but also a capability to discover new motif sites in microorganisms. ECPICK is a reliable EC number prediction tool to identify protein functions of an increasing number of uncharacterized enzymes.

GjSOX9 Cloning, Expression, and Comparison with gjSOXs Family Members in Gekko japonicus.

SOX9 plays a crucial role in the male reproductive system, brain, and kidneys. In this study, we firstly analyzed the complete cDNA sequence and expression patterns for SOX9 from Gekko japonicus SOX9 (gjSOX9), carried out bioinformatic analyses of physiochemical properties, structure, and phylogenetic evolution, and compared these with other members of the gjSOX family. The results indicate that gjSOX9 cDNA comprises 1895 bp with a 1482 bp ORF encoding 494aa. gjSOX9 was not only expressed in various adult tissues but also exhibited a special spatiotemporal expression pattern in gonad tissues. gjSOX9 was predicted to be a hydrophilic nucleoprotein with a characteristic HMG-Box harboring a newly identified unique sequence, "YKYQPRRR", only present in SOXE members. Among the 20 SOX9 orthologs, gjSOX9 shares the closest genetic relationships with Eublepharis macularius SOX9, Sphacrodactylus townsendi SOX9, and Hemicordylus capensis SOX9. gjSOX9 and gjSOX10 possessed identical physicochemical properties and subcellular locations and were tightly clustered with gjSOX8 in the SOXE group. Sixteen gjSOX family members were divided into six groups: SOXB, C, D, E, F, and H with gjSOX8, 9, and 10 in SOXE among 150 SOX homologs. Collectively, the available data in this study not only facilitate a deep exploration of the functions and molecular regulation mechanisms of the gjSOX9 and gjSOX families in G. japonicus but also contribute to basic research regarding the origin and evolution of SOX9 homologs or even sex-determination mode in reptiles.

Mining Top-k High Average-Utility Sequential Patterns for Resource Transformation

High-utility sequential pattern mining (HUSPM) helps researchers find all subsequences that have high utility in a quantitative sequential database. The HUSPM approach appears to be well suited for resource transformation in DIKWP graphs. However, all the extensions of a high-utility sequential pattern (HUSP) also have a high utility that increases with its length. Therefore, it is difficult to obtain diverse patterns of resources. The patterns that consist of many low-utility items can also be a HUSP. In practice, such a long pattern is difficult to analyze. In addition, the low-utility items do not always reflect the interestingness of association rules. High average-utility pattern mining is considered a solution to extract more significant patterns by considering the lengths of patterns. In this paper, we formulate the problem of top-k high average-utility sequential pattern mining (HAUSPM) and propose a novel algorithm for resource transformation. We adopt a projection mechanism to improve efficiency. We also adopt the sequence average-utility-raising strategy to increase thresholds. We design the prefix extension average utility and the reduced sequence average utility by incorporating the average utility into the utility upper bounds. The results of our comparative experiments demonstrate that the proposed algorithm can achieve sufficiently good performance.

Children With Developmental Language Disorder Show Deficits in the Production of Musical Rhythmic Groupings.

Children with developmental language disorder (DLD) show evidence of domain-general deficits in sequentially patterned motor skills. This study focuses on the production of rhythmically grouped sequences drawn from a music task, with the hypothesis that children with DLD will show a sequential pattern learning deficit that crosses language and action domains. Fifty-seven 4- to 5-year-old children (36 with DLD) drummed and clapped a developmentally appropriate musical rhythmic sequence 24 times (clapped 12 times, drummed 12 times). The accuracy of rhythmic events (markings of claps, drums, and pauses in a target sequence) was assessed through a modification of classic speech and language transcription procedures. The variability and prosodic structure of the rhythmic groupings were also measured. Children with DLD produced less accurate and more variable rhythmic groupings compared to their typically developing (TD) peers. While the final-position grouping of the sequence was especially vulnerable for all children, those with DLD included more co-occurring errors in initial and final groupings of the same rhythmic sequence. Both TD children and children with DLD were less accurate in the clapping than the drumming task. Neither rhythmic drumming nor clapping accuracy correlated with motor skill in either group of children. This study provides novel evidence of a manual rhythmic grouping deficit in DLD, one that is motivated by language-not motor or speech-factors. Cognitive abilities necessary to organize rhythmic events into higher order groupings are impaired across music and language in children with DLD. Rhythmic organization and sequencing may serve an important role in diagnosis and intervention in this population. https://doi.org/10.23641/asha.24158745.

Charge Distribution Patterns of IA3 Impact Conformational Expansion and Hydration Diffusivity of the Disordered Ensemble.

IA3 is a 68 amino acid natural peptide/protein inhibitor of yeast aspartic proteinase A (YPRA) that is intrinsically disordered in solution with induced N-terminal helicity when in the protein complex with YPRA. Based on the intrinsically disordered protein (IDP) parameters of fractional net charge (FNC), net charge density per residue (NCPR), and charge patterning (κ), the two domains of IA3 are defined to occupy different domains within conformationally based subclasses of IDPs, thus making IA3 a bimodal domain IDP. Site-directed spin labeling (SDSL) electron paramagnetic resonance (EPR) spectroscopy and low-field Overhauser dynamic nuclear polarization (ODNP) spectroscopy results show that these two domains possess different degrees of compaction and hydration diffusivity behavior. This work suggests that SDSL EPR line shapes, analyzed in terms of their local tumbling volume (VL), provide insights into the compaction of the unstructured IDP ensemble in solution and that protein sequence and net charge distribution patterns within a conformational subclass can impact bound water hydration dynamics, thus possibly offering an alternative thermodynamic property that can encode conformational binding and behavior of IDPs and liquid-liquid phase separations.

BERD+: A Generic Sequential Recommendation Framework by Eliminating Unreliable Data with Item- and Attribute-level Signals

Most sequential recommendation systems (SRSs) predict the next item as the target for users given its preceding items as input, assuming the target is definitely related to its input. However, users may unintentionally click items that are inconsistent with their preference due to external factors, causing unreliable instances whose target mismatches the input. We, for the first time , verify SRSs can be misguided by such unreliable instances and design a generic SRS framework B y E liminating un R eliable D ata (BERD+), which can be flexibly plugged into existing SRSs. Specifically, BRED+ is guided with observations on the training process of instances: Unreliable instances generally have high training loss; high-loss instances are not necessarily unreliable but uncertain ones caused by blurry sequential patterns; and item attributes help rectify instance loss and uncertainty, but may also introduce disturbance. Accordingly, BERD+ models both the loss and uncertainty of each instance via a Gaussian distribution, whereby a heterogeneous uncertainty-aware graph convolution network is designed to learn accurate embeddings for different entities while reducing the disturbance caused by uncertain attribute values. Thereafter, an explicit preference extractor rectifies instance loss and uncertainty and reduces the disturbance caused by less-focused attribute types. Finally, instances with high loss and low uncertainty are eliminated as unreliable data. Extensive experiments verify the efficacy of BERD+.

DA-DAN: A Dual Adversarial Domain Adaption Network for Unsupervised Non-overlapping Cross-domain Recommendation

Unsupervised Non-overlapping Cross-domain Recommendation (UNCR) is the task that recommends source domain items to the target domain users, which is more challenging as the users are non-overlapped, and its learning process is unsupervised. Unsupervised Non-overlapping Cross-domain Recommendation UNCR is still unsolved due to the following: (1) Previous studies need extra auxiliary information to learn transferable features when aligning two domains, which is unrealistic and hard to obtain due to privacy concerns. (2) Since the adoption of the shared network, existing works cannot well eliminate the domain-specific features in the common feature space, which may incorporate domain noise and harm the cross-domain recommendation. In this work, we propose a domain adaption-based method, namely DA-DAN, to address the above challenges. Specifically, to let DA-DAN be free of auxiliary information, we learn users’ preferences by only exploring their sequential patterns, and propose an improved self-attention layer to model them. To well eliminate the domain-specific features from the common feature space, we resort to a dual generative adversarial network with a multi-target adversarial loss, where two generators and discriminators are leveraged to model each domain separately. Experimental results on three real-world datasets demonstrate the advantage of DA-DAN compared with the state-of-the-art recommendation baselines. Moreover, our source codes have been publicly released. 1

Collaborative analytics-enhanced reflective assessment to foster conducive epistemic emotions in knowledge building

Fostering conducive epistemic emotions to students' knowledge building (KB) is critical but challenging. Considering the potential role of metacognition and KB in promoting conducive epistemic emotions, we designed a quasi-experiment and examined the influence of collaborative analytics-enhanced reflective assessment (demonstrate metacognition) on conducive epistemic emotions to students' KB. The experimental class included 48 undergraduates engaged in KB inquiries enhanced by collaborative analytics-enhanced reflective assessment. In contrast, the comparison class included 46 undergraduates engaged in KB enhanced by portfolio assessment. Students' online Knowledge Forum discourse was collected and coded regarding the demonstrated epistemic emotions. Comparative analysis of the frequency distribution and lag sequential patterns of epistemic emotions revealed that the collaborative analytics-enhanced reflective assessment design fostered conducive epistemic emotions to KB among the undergraduates. Correlation analysis revealed positive correlations between conducive epistemic emotions, higher-level KB discourse moves, and individual knowledge creation. The findings enrich our understanding of promoting undergraduates' conducive epistemic emotions and uncover the relationships between metacognition, epistemic emotions, and KB practices. It provides implications for developing a reflective and collaborative culture, providing scaffoldings and opportunities for expressing and analyzing epistemic emotions, utilizing prompt sheets and collective reflections to enhance the productive use of analytics, and fostering a conducive emotional experience for students.

Discovering patterns in outpatient neurology appointments using state sequence analysis

BackgroundOutpatient services in the UK, and in particular outpatient neurology services, are under considerable pressure with an ever-increasing gap between capacity and demand. To improve services, we first need to understand the current situation. This study aims to explore the patterns of appointment type seen in outpatient neurology, in order to identify potential opportunities for change.MethodsWe use State Sequence Analysis (SSA) on routinely collected data from a single neurology outpatient clinic. SSA is an exploratory methodology which allows patterns within sequences of appointments to be discovered. We analyse sequences of appointments for the 18 months following a new appointment. Using SSA we create groups of similar appointment sequence patterns, and then analyse these clusters to determine if there are particular sequences common to different diagnostic categories.ResultsOf 1315 patients 887 patients had only one appointment. Among the 428 patients who had more than one appointment a 6 monthly cycle of appointments was apparent. SSA revealed that there were 11 distinct clusters of appointment sequence patterns. Further analysis showed that there are 3 diagnosis categories which have significant influence over which cluster a patient falls into: seizure/epilepsy, movement disorders, and headache.ConclusionsNeurology outpatient appointment sequences show great diversity, but there are some patterns which are common to specific diagnostic categories. Information about these common patterns could be used to inform the structure of future outpatient appointments.

MR-AT: Map Reduce based Apriori Technique for Sequential Pattern Mining using Big Data in Hadoop

One of the most well-known and widely implemented data mining methods is Apriori algorithm which is responsible for mining frequent item sets. The effectiveness of the Apriori algorithm has been improved by a number of algorithms that have been introduced on both parallel and distributed platforms in recent years. They are distinct from one another on account of the method of load balancing, memory system, method of data degradation, and data layout that was utilised in their implementation. The majority of the issues that arise with distributed frameworks are associated with the operating costs of handling distributed systems and the absence of high-level parallel programming languages. In addition, when using grid computing, there is constantly a possibility that a node will fail, which will result in the task being re-executed multiple times. The MapReduce approach that was developed by Google can be used to solve these kinds of issues. MapReduce is a programming model that is applied to large-scale distributed processing of data on large clusters of commodity computers. It is effective, scalable, and easy to use. MapReduce is also utilised in cloud computing. This research paper presents an enhanced version of the Apriori algorithm, which is referred to as Improved Parallel and Distributed Apriori (IPDA). It is based on the scalable environment referred as Hadoop MapReduce, which was used to analyse Big Data. Through the generation of split-frequent data regionally and the early elimination of unusual data, the proposed work has its primary objective to reduce the enormous demands placed on available resources as well as the reduction of the overhead communication that occurs whenever frequent data are retrieved. The paper presents the results of tests, which demonstrate that the IPDA performs better than traditional apriori and parallel and distributed apriori in terms of the amount of time required, the number of rules created, and the various minimum support values.

A survey of high utility sequential patterns mining methods

In recent years, there has been an increasing demand for high utility sequential pattern (HUSP) mining. Different from high utility itemset mining, the “combinatorial explosion” problem of sequence data makes it more challenging. This survey aims to provide a general, comprehensive, and structured overview of the state-of-the-art methods of HUSP from a novel perspective. Firstly, from the perspective of serial and parallel, the data structure used by the mining methods are illustrated and the pros and cons of the algorithms are summarized. In order to protect data privacy, many HUSP hiding algorithms have been proposed, which are classified into array-based, chain-based and matrix-based algorithms according to the key technologies. The hidden strategies and evaluation metrics adopted by the algorithms are summarized. Next, a taxonomy of the most common and the state-of-the-art approaches for incremental mining algorithms is presented, including tree-based and projection-based. In order to deal with the latest sequence in the data stream, the existing algorithms often use the window model to update dynamically, and the algorithms are divided into methods based on sliding windows and landmark windows for analysis. Afterwards, a summary of derived high utility sequential pattern is presented. Finally, aiming at the deficiencies of the existing HUSP research, the next work that the author plans to do is given.

CCSMP: an efficient closed contiguous sequential pattern mining algorithm with a pattern relation graph

CCSMP: an efficient closed contiguous sequential pattern mining algorithm with a pattern relation graph

Pulmonary disease detection and classification in patient respiratory audio files using long short-term memory neural networks.

In order to improve the diagnostic accuracy of respiratory illnesses, our research introduces a novel methodology to precisely diagnose a subset of lung diseases using patient respiratory audio recordings. These lung diseases include Chronic Obstructive Pulmonary Disease (COPD), Upper Respiratory Tract Infections (URTI), Bronchiectasis, Pneumonia, and Bronchiolitis. Our proposed methodology trains four deep learning algorithms on an input dataset consisting of 920 patient respiratory audio files. These audio files were recorded using digital stethoscopes and comprise the Respiratory Sound Database. The four deployed models are Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), CNN ensembled with unidirectional LSTM (CNN-LSTM), and CNN ensembled with bidirectional LSTM (CNN-BLSTM). The aforementioned models are evaluated using metrics such as accuracy, precision, recall, and F1-score. The best performing algorithm, LSTM, has an overall accuracy of 98.82% and F1-score of 0.97. The LSTM algorithm's extremely high predictive accuracy can be attributed to its penchant for capturing sequential patterns in time series based audio data. In summary, this algorithm is able to ingest patient audio recordings and make precise lung disease predictions in real-time.

KitBit: A New AI Model for Solving Intelligence Tests and Numerical Series.

The resolution of intelligence tests, in particular numerical sequences, has been of great interest in the evaluation of AI systems. We present a new computational model called KitBit that uses a reduced set of algorithms and their combinations to build a predictive model that finds the underlying pattern in numerical sequences, such as those included in IQ tests and others of much greater complexity. We present the fundamentals of the model and its application in different cases. First, the system is tested on a set of number series used in IQ tests collected from various sources. Next, our model is successfully applied on the series used to evaluate the models reported in the literature. In both cases, the system is capable of solving these types of problems in less than a second using standard computing power. Finally, KitBit's algorithms have been applied for the first time to the complete set of entire series of the well-known OEIS database. We find a pattern in the form of a list of algorithms and predict the following terms in the largest number of series to date. These results demonstrate the potential of KitBit to solve complex problems that could be represented numerically.

Detecting visual texture patterns in binary sequences through pattern features.

We measured human ability to detect texture patterns in a signal detection task. Observers viewed sequences of 20 blue or yellow tokens placed horizontally in a row. They attempted to discriminate sequences generated by a random generator ("a fair coin") from sequences produced by a disrupted Markov sequence (DMS) generator. The DMSs were generated in two stages: first a sequence was generated using a Markov chain with probability, pr = 0.9, that a token would be the same color as the token to its left. The Markov sequence was then disrupted by flipping each token from blue to yellow or vice versa with probability, pd-the probability of disruption. Disruption played the role of noise in signal detection terms. We can frame what observers are asked to do as detecting Markov texture patterns disrupted by noise. The experiment included three conditions differing in pd (0.1, 0.2, 0.3). Ninety-two observers participated, each in only one condition. Overall, human observers' sensitivities to texture patterns (d' values) were markedly less than those of an optimal Bayesian observer. We considered the possibility that observers based their judgments not on the entire texture sequence but on specific features of the sequences such as the length of the longest repeating subsequence. We compared human performance with that of multiple optimal Bayesian classifiers based on such features. We identify the single- and multiple-feature models that best match the performance of observers across conditions and develop a pattern feature pool model for the signal detection task considered.

TERI: An Effective Framework for Trajectory Recovery with Irregular Time Intervals

The proliferation of trajectory data has facilitated various applications in urban spaces, such as travel time estimation, traffic monitoring, and flow prediction. These applications require a substantial volume of high-quality trajectories as the prerequisite to achieve effective performance. Unfortunately, a large number of real-world trajectories are inevitably collected in unsatisfactory quality due to device constraints. To address this issue, previous studies have proposed numerous trajectory recovery methods to augment the quality of such trajectories, thereby ensuring the performance of related applications. However, these methods all assume the awareness of the recovery positions in advance, which is a condition not always available in practice. In this paper, we discard this strong assumption and focus on trajectory recovery with irregular time intervals as a more prevalent setting in downstream scenarios. We propose a novel framework, called TERI, to tackle trajectory recovery without prior information in a two-stage process, where recovery positions are first detected, followed by the imputation of the missing data points. In each stage, TERI framework deploys a model named RETE, which is based on Transformer encoder architecture enhanced by novel designs to boost the performance for the new problem setting. Specifically, RETE features a learnable Fourier encoding module to better model spatial and temporal correlations, and integrates collective transition pattern learning and trajectory contrastive learning to effectively capture sequential transition patterns. Extensive experiments on three real-world datasets demonstrate that TERI consistently outperforms all the baselines by a significant large margin.

Searching for new early detection markers of oral epithelial dysplasia and oral squamous cell carcinoma using oral liquid-based cytology

ObjectiveOral squamous cell carcinomas (SCCs) arise from oral potentially malignant disorders (OPMDs), including oral epithelial dysplasia (OED); however, useful detections markers for early detection of OED and oral SCC are lacking. In this study, we aimed to evaluate the sequential mRNA and protein expression patterns of BRD4, c-MYC, TP53, and MUC21 in human oral samples by liquid-based cytology (LBC) and determine whether these markers are useful for early detection of OED and oral SCC when combined with oral cytological screening. MethodsA total 200 patients were obtained from our hospital between 2020 and 2022. These are samples from same cases with oral cytology and corresponding histological diagnosis. The expression of these markers was examined by immunocytochemistry and mRNA analysis. The percentages of cells with nuclear or cytoplasmic immunostaining were calculated using the labeling index (LI). BRD4, c-MYC, and TP53 mRNA expression levels were upregulated in tissue during progression from a status of negative for intraepithelial lesions or malignancy (NILM) to positive for SCC. BRD4-LI and c-MYC-LI significantly increased in low-grade squamous intraepithelial lesions (LSIL), high-grade squamous intraepithelial lesions (HSIL) and SCC samples, whereas the expression of MUC21 mRNA and MUC21-LI gradually decreased. TP53-LI increased only in SCC samples. ConclusionsWe conclude that monitoring BRD4 and c-MYC expression with immunocytochemistry could improve the early detection of OED and oral SCC.