t-SNE Dimensionality Reduction Research Articles

Classification of soybean chemical characteristics by excitation emission matrix coupled with t-SNE dimensionality reduction

Measuring the chemical composition in soybeans is time-consuming and laborious, and even simple near-infrared sensors generally require the creation of calibration curves before application. In this study, a new screening method for soybeans without calibration curves was investigated by combining the excitation emission matrix (EEM) and dimensionality reduction analysis. The EEMs of 34 soybean samples were measured, and representative chemical contents including crude protein, crude oil and isoflavone contents were measured by chemical analysis. Two methods of dimensionality reduction: principal component analysis (PCA) and t-distributed Stochastic Neighbor Embedding (t-SNE) were applied on the EEM data to obtain two-dimensional plots, which were divided into two regions with large or small amount of each chemical components. To classify the large or small levels of each of the chemical composition, machine learning classification models were constructed on the two-dimensional plots after dimensionality reduction. As a result, the classification accuracy was higher in t-SNE than in the combinations of PC1 and PC2 from PCA. Furthermore, in t-SNE, the classification accuracy reached over 90% for all the chemical components. From these results, t-SNE dimensionality reduction on the soybean EEM has the potential for easy and accurate screening of soybeans especially based on isoflavone contents.

Ensemble Meta-Learning-Based Robust Chipping Prediction for Wafer Dicing

Our previous study utilized importance analysis, random forest, and Barnes–Hut t-SNE dimensionality reduction to analyze critical dicing parameters and used bidirectional long short-term memory (BLSTM) to predict wafer chipping occurrence successfully in a single dicing machine. However, each dicing machine of the same type may produce unevenly distributed non-IID dicing signals, which may lead to the undesirable result that a pre-trained model trained by dicing machine #1 could not effectively predict chipping occurrence in dicing machine #2. Therefore, regarding the model robustness, this study introduces an ensemble meta-learning-based model that can evaluate many dicing machines for chipping prediction with high stability and accuracy. This approach constructs several base learners, such as the hidden Markov model (HMM), the variational autoencoder (VAE), and BLSTM, to form an ensemble learning. We use model-agnostic meta-learning (MAML) to train and test the ensemble learning model by several prediction tasks from machine #1. After MAML learning, we call the trained model a meta learner. Then, we successfully apply a retrieved data set from machine #2 to the meta learner for training and testing wafer chipping occurrence in this machine. As a result, our contribution to the robust chipping prediction on cross-machines can improve the yield of wafer dicing with a prediction accuracy of 93.21%, preserve the practical wearing of dicing kerfs, and significantly cut wafer manufacturing costs.

Integrating Traditional Machine Learning and Deep Learning for Precision Screening of Anticancer Peptides: A Novel Approach for Efficient Drug Discovery.

The rapid and effective identification of anticancer peptides (ACPs) by computer technology provides a new perspective for cancer treatment. In the identification process of ACPs, accurate sequence encoding and effective classification models are crucial for predicting their biological activity. Traditional machine learning methods have been widely applied in sequence analysis, but deep learning provides a new approach to capture sequence complexity. In this study, a two-stage ACPs classification model was innovatively proposed. Three novel coding strategies were explored; two mainstream Natural Language Processing (NLP) models and 11 machine learning models were fused to identify ACPs, which significantly improved the prediction accuracy of ACPs. We analyzed the correlation between peptide chain amino acids and evaluated the relevant performance of the model by the ROC curve and t-SNE dimensionality reduction technique. The results indicated that the deep learning and machine learning fusion models of M3E-base and KNeighborsDist models, especially when considering the semantic information on amino acid sequences, achieved the highest average accuracy (AvgAcc) of 0.939, with an AUC value as high as 0.97. Then, in vitro cell experiments were used to verify that the two ACPs predicted by the model had antitumor efficacy. This study provides a convenient and effective method for screening ACPs. With further optimization and testing, these strategies have the potential to play an important role in drug discovery and design.

Unsupervised Multivariate Feature-Based Adaptive Clustering Analysis of Epileptic EEG Signals.

Supervised classification algorithms for processing epileptic EEG signals rely heavily on the label information of the data, and existing supervised methods cannot effectively solve the problem of analyzing unlabeled epileptic EEG signals. In the traditional unsupervised clustering algorithm, the number of clusters and the global parameters must be predetermined, and the algorithm's analytical results are combined with a huge number of subjective errors, which affects the detection accuracy. For this reason, this paper proposes an unsupervised multivariate feature adaptive clustering analysis algorithm based on epileptic EEG signals. First, CEEMDAN and CWT are introduced into the epileptic EEG signal after preprocessing for joint denoising to further improve the signal quality. Then, the multivariate feature set of the signal is extracted and constructed, which includes nonlinear, time, frequency, and time-frequency characteristics. To reveal the hidden structures and correlations in the high-dimensional feature data, t-SNE dimensionality reduction is introduced. Finally, the DBSCAN clustering algorithm is optimized using the SSA algorithm to achieve adaptive selection of cluster number and global parameters.It not only enhances the clustering performance and reliability of the clustering results, but also avoids subjective errors in the analysis results. It provides a pre-theoretical foundation for the successful development of future seizure prediction devices and has good application prospects in clinical diagnosis and daily monitoring of patients.

Establishment and Application of Steel Composition Prediction Model Based on t-Distributed Stochastic Neighbor Embedding (t-SNE) Dimensionality Reduction Algorithm

Establishment and Application of Steel Composition Prediction Model Based on t-Distributed Stochastic Neighbor Embedding (t-SNE) Dimensionality Reduction Algorithm

Face extraction and clustering with Viola - Jones object detection framework and T-SNE dimensionality reduction

We investigate the possibility to use Viola-Jones [1] object detection framework through a multi-model approach to build a face extraction pipeline that will be used in video appearance tagging. Although deep convolutional neural networks have surpassed previous algorithms in performance [2], Haar Cascades needs much lower memory than CNN, does not require specialized hardware, and has lower storage requirements. Most videos will show the same face more than once, at least a few close-ups that are full frontal and well lit. We need an efficient system that will extract the best appearances. This study shows the pre-trained model selection, the fine-tuning of run-time parameters and the test. After selection of models for faces, eyes, mouths and noses and testing the right runtime parameters we were able to establish a procedure that will avoid any false positives and will produce a set of well defined faces.tart your abstract here

GHGPR-PPIS: A graph convolutional network for identifying protein-protein interaction site using heat kernel with Generalized PageRank techniques and edge self-attention feature processing block

Accurately pinpointing protein-protein interaction site (PPIS) on the molecular level is of utmost significance for annotating protein function and comprehending the mechanisms underpinning various diseases. While numerous computational methods for predicting PPIS have emerged, they have indeed mitigated the labor and time constraints associated with traditional experimental methods. However, the predictive accuracy of these methods has yet to reach the desired threshold. In this context, we proposed a groundbreaking graph-based computational model called GHGPR-PPIS. This innovative model leveraged a graph convolutional network using heat kernel (GraphHeat) in conjunction with Generalized PageRank techniques (GHGPR) to predict PPIS. Additionally, building upon the GHGPR framework, we devised an edge self-attention feature processing block, further augmenting the performance of the model. Experimental findings conclusively demonstrated that GHGPR-PPIS surpassed all competing state-of-the-art models when evaluated on the benchmark test set. Impressively, on two distinct independent test sets and a specific protein chain, GHGPR-PPIS consistently demonstrated superior generalization performance and practical applicability compared to the comparative model, AGAT-PPIS. Lastly, leveraging the t-SNE dimensionality reduction algorithm and clustering visualization technique, we delved into an interpretability analysis of the effectiveness of GHGPR-PPIS by meticulously comparing the outputs from different stages of the model.

Detection of Ocean Internal Waves Based on Modified Deep Convolutional Generative Adversarial Network and WaveNet in Moderate Resolution Imaging Spectroradiometer Images

The generation and propagation of internal waves in the ocean are a common phenomenon that plays a pivotal role in the transport of mass, momentum, and energy, as well as in global climate change. Internal waves serve as a critical component of oceanic processes, contributing to the redistribution of heat and nutrients in the ocean, which, in turn, has implications for global climate regulation. However, the automatic identification of internal waves in oceanic regions from remote sensing images has presented a significant challenge. In this research paper, we address this challenge by designing a data augmentation approach grounded in a modified deep convolutional generative adversarial network (DCGAN) to enrich MODIS remote sensing image data for the automated detection of internal waves in the ocean. Utilizing t-distributed stochastic neighbor embedding (t-SNE) technology, we demonstrate that the feature distribution of the images produced by the modified DCGAN closely resembles that of the original images. By using t-SNE dimensionality reduction technology to map high-dimensional remote sensing data into a two-dimensional space, we can better understand, visualize, and analyze the quality of data generated by the modified DCGAN. The images generated by the modified DCGAN not only expand the dataset’s size but also exhibit diverse characteristics, enhancing the model’s generalization performance. Furthermore, we have developed a deep neural network named “WaveNet,” which incorporates a channel-wise attention mechanism to effectively handle complex remote sensing images, resulting in high classification accuracy and robustness. It is important to note that this study has limitations, such as the reliance on specific remote sensing data sources and the need for further validation across various oceanic regions. These limitations are essential to consider in the broader context of oceanic research and remote sensing applications. We initially pre-train WaveNet using the EuroSAT remote sensing dataset and subsequently employ it to identify internal waves in MODIS remote sensing images. Experiments show the highest average recognition accuracy achieved is an impressive 98.625%. When compared to traditional data augmentation training sets, utilizing the training set generated by the modified DCGAN leads to a 5.437% enhancement in WaveNet’s recognition rate.

Dimensionality reduction for detection of anomalies in the IoT traffic data

This paper concerns cybersecurity issues in one of the fastest growing fields of modern computer systems the Internet-of-Things (IoT). In this field, intrusion detection plays a significant role in allowing IoT systems’ safe functioning and continuous operation. Visualizing the IoT data using dimensionality reduction allows for an easy and straightforward traffic analysis based on a graphical data representation, eligible for interpretation even by non-experts. We present a study on IoT network intrusion detection using three dimensionality reduction methods, namely, the Self-Organizing Map (SOM), the t-distributed Stochastic Neighbor Embedding (t-SNE), and the Neighborhood Retrieval Visualizer (NeRV). We show that applying them to the IoT traffic allows for reducing the original traffic feature space to a 2-D one, where anomalies may be noticed visually as outliers. The purpose of our study and its original contribution is conducting a comparative analysis of the t-SNE versus NeRV dimensionality reduction approaches in both: theoretical and empirical aspects. We notice and point out specific significant differences between these methods, which, as we claim, are responsible for their different performance in the IoT field, which is validated by our empirical study on real-world IoT traffic datasets. The results of our experimental research provide an interesting insight into the behavior of the investigated techniques and confirm their effectiveness and usability in IoT anomalies detection.

Single-cell analysis of the miRNA activities in tuberculous meningitis (TBM) model mice injected with the BCG vaccine

BackgroundOur previous study revealed the transcriptome atlas of specific cell types in tuberculous meningitis (TBM) model mice injected with the BCG vaccine via scRNA sequencing. However, the activities of miRNAs in TBM at single-cell resolution remain to be explored. MethodCell type-specific miRNA activities were investigated by using motif enrichment analyses (miReact) on the transcriptome data of 15 cell types. The target mRNAs of miRNAs were predicted and subjected to enrichment analysis. Furthermore, miRNAs and their target mRNAs with opposite expression trends were chosen to construct functional networks. Besides, qRT-PCR and RNA scope were performed to verify the expression level of representative miRNA. ResultsThe tSNE dimensionality reduction presented 15 cell types in TBM model mice, in which microglia and endothelial cells accounted for the majority. Target mRNAs of each cell type were predicted for verification or network construction. The immune and inflammation-related miRNA-mRNA networks of macrophages and microglia, oxidative phosphorylation-related miRNA-mRNA networks of neurons, ion and protein transport-related networks of epididymal cells, and angiogenesis-related miRNA-mRNA networks of VSMCs were constructed. The miRNA activity analysis revealed that miR-21a-3p activity was increased in microglia, macrophages, neurons and epididymal cells. The result of qRT-PCR and RNA scope indicate that miR-21a-3p was significantly higher-expressed in TBM brain tissue compared with normal brain tissue. ConclusionIn our study, an in-depth exploration of the mRNA expression and miRNA activity of macrophages, microglia, epididymal cells, neurons and vascular smooth muscle cells during TBM progression was conducted using scRNA-Seq, which provided novel insights into the immune cell engagement in TBM patients.

P05.03.B METHYLOME PROFILING OF GLIOBLASTOMAS WITH FGFR3/TACC3 FUSION

Abstract BACKGROUND Glioblastomas (GBMs) with the FGFR3-TACC3 (F3T3) gene fusion have histological, molecular, metabolic, and clinicoradiological characteristics. They have a better prognosis compared to other GBMs. Recent findings suggest that methylome profiling could further stratify these patients identifying a subgroup (defined as “outlier” F3T3+ GBMs, or GBM-F3T3-O) with a methylation profile resembling gangliogliomas (GGs), low-grade histological appearance, and a better outcome. MATERIAL AND METHODS We performed DNA methylation profiling (Infinium MethylationEPIC Array v1.0) of 32 adult diffuse gliomas with FGFR3 gene fusions (30 FGFR3-TACC3, one FGFR3-ACLY, and one FGFR3-MYO18A), plus two paired samples of a diffuse glioma with two distinct contingents (one with positive FGFR3 staining and FGFR3-TACC3 gene fusion, the other with intense EGFR staining and EGFR gene amplification). All cases corresponded to “glioblastoma, IDH-wildtype” according to the WHO CNS5. Methylation data were submitted to the DFKZ brain tumor classifier (molecularneuropathology.org), versions 11b4 and 12.5 and to dimensionality reduction with t-distributed stochastic neighbor embedding (tSNE). Copy-number variation (CNV) estimation based on conumee package was obtained from DFKZ website. RESULTS With the brain tumor classifier v11b4, most cases (26/34, 76%) were classified as methylation class (MC) GBM (17/26 with a significant score ≥0.9). Out of them, 23/26 (88%) were significantly (score ≥0.5) classified in the “GBM, mesenchymal” (16/26, 62%) or in the “GBM, RTK II” methylation families (7/26, 27%). With the classifier v12.5, again most tumors were classified as “GBM, mesenchymal” (18/34, 53%) or “GBM, RTK II” (9/34, 26%), of which twelve (12/34, 35%) with a score ≥0.9. Two were classifed as GG (score 0.69 and 0.39), both having “GBM, mesenchymal” as a second hit (score 0.19 and 0.33, respectively). Five had GG as a second hit, although with low scores (range <0.01-0.25). After tSNE dimensionality reduction of our and literature cases, most tumors clustered together with other GBM, mesenchymal and GBM, RTK II cases. Only two cases fell within a distinct cluster corresponding to GBM-F3T3-O tumors from the literature. These included two histologically defined GBMs with a survival of 26 and 100 months. Recurrent inferred CNV included chromosome (chr) 7 gain (96%) and chr10 loss (85%), followed by chr20 gain (46%), chr22q loss (46%), and chr19 gain (33%). The presence of chr19 gain (alone or associated with chr20 gain) was associated with increased survival (p=0.01). CONCLUSION Glioblastomas with FGFR3 fusions (mostly F3T3) have an epigenomic profile close to other mesenchymal and RTK II GBMs. Methylation analysis identifies methylation (GBM-F3T3-O) and CNV (chr19 gain) profiles associated with better survival.

Взаимосвязь показателей «предстартового» состояния спортсмена с физиологической реакцией на стандартизированную нагрузку

Intense physical work is characterized by activity of physiological mechanisms as interrelated components joint for physical exertion. Definition of a set of individual and typological patterns of the physiological mechanisms’ activity answers the questions related to improvement of the athlete’s potential realization efficiency, definition of the limiting components and body’s reserve capacity, training load management. The study was aimed to assess the relationship between the responses of physiological mechanisms associated with standardized physical exertion and the pre-start state parameters. The athlete was through the step incremental test with the treadmill involving recording of the gas exchange parameters and heart rate to study physiological patterns. The physiological response parameters were calculated relative to the key phases of the exercise test: pre-start state, aerobic and anaerobic thresholds, peak exertion, fast and slow recovery phases. The mathematical model “Horseshoe of Rest” characterizing the athlete’s pre-start state before performing the test was constructed using the T-SNE dimensionality reduction algorithms. The model enables estimation of the release of non-metabolic СО2 throughout the testing period (MIC — 0.29) and the exertion period (MIC — 0.35).

Paper-based multiplexed colorimetric biosensing of cardiac and lipid biomarkers integrated with machine learning for accurate acute myocardial infarction early diagnosis and prognosis

This study demonstrates how a colorimetric biosensor based on microfluidic paper can swiftly diagnose a disease and predict its prognosis to triage patients effectively. This was the first biosensor to quantify the gold standard cardiac troponin (cTnI) and lipid biomarkers, including high-density lipoprotein (HDL) and low-density lipoprotein (LDL) simultaneously. Prior research encountered obstacles or limitations, such as measuring a single biomarker or total cholesterol, which cannot distinguish between LDL and HDL. CatBoost, an advanced machine learning (ML) technique used for diagnosis that combines the predictive power of ML algorithms obtained an impressive area under the receiver operating curves (AUROC) of 0.97 ± 0.018 for all possible classification thresholds. CatBoost is a brand-new ensemble framework based on the interaction of multiple health parameters that can generate AUROC values of 0.897 ± 0.047 for the accurate prognosis of recurrent acute myocardial infarction (AMI), demonstrating remarkably accurate AMI diagnosis and prognosis. In addition, this paper-based analytical device (µPAD) biosensor employs an electrophoretic method to overcome the challenges posed by non-specific adsorption. This is accomplished by isolating non-specific biomolecules based on differences in their isoelectric points and removing non-specifically adsorbing colorimetric markers. The limits of detection (LoD) for cTnI in AMI were lower than their respective clinical cutoff values. This study also demonstrated that the proposed ML framework produced significantly better results than conventional statistical analysis. High correlation filtering and (t-SNE) dimensionality reduction were utilized for a limited number of data points. The respectable accuracy and AUROC of this method were also validated using cross-validation.

P12 Unsupervised Clustering and Binary Classification Analysis for Predicting COVID Syndrome (LONG COVID) in US Adults

This study investigated the risk factors of developing COVID Syndrome and identified potential disease profiles that may exist among those who have contracted COVID-19. Data on 13,953 adults who had experienced COVID-19 at any time were analyzed from the 2022 US National Health and Wellness Survey. XGBoost binary classification with 10-fold cross-validation was used to predict long COVID among those who reported experiencing COVID-19 and to extract feature importance. Synthetic minority oversampling technique (SMOTE) was used to address class imbalance in the outcome variable. Variable selection was conducted based on SHAP values. Fifty variables including demographic characteristics, COVID-19 symptoms, comorbidities, and health characteristics were used in the final model. Parameters were tuned using AUC. Among the 2,665 respondents who were diagnosed with long COVID, k-medoids clustering with t-SNE dimensionality reduction was implemented to determine whether distinct symptom profiles exist. Average silhouette score was used to determine the optimal number of clusters. The XGBoost binary classification for predicting long COVID among those with COVID-19 had an AUC of 0.9145, accuracy of 0.9072, sensitivity of 0.9630, specificity of 0.8328, and Brier score of 0.0928. The most important features in predicting long COVID were age, smoking habits, COVID-19 vaccination status, certain COVID-19 symptoms experienced, and certain comorbidities. Among those diagnosed with long COVID, the clustering analysis found nine unique clusters of symptoms. The cluster that experienced the most severe symptoms was older, female, lower income, lower vaccination rate, and had more comorbidities like asthma, chronic bronchitis, and allergies. In a broadly representative US adult population, XGBoost model identified a selection of risk factors for developing long COVID. K-medoids clustering identified clusters of patients that were at risk for developing severe symptoms.

Non-fatal outcomes of COVID-19 disease in children with solid organ transplants associates with down-regulation of senescent pathways

Abstract Objective: To study the immune landscape during acute infection (AI) & COVID-19 recovery in children with solid organ transplants (SOTs). Method: Blood collected between Nov 2020 – June 2022 from pediatric SOT recipients 1–14-days after a positive PCR for SARS-CoV-2 (AI), & 15–60-days after a positive PCR (convalescence), for Ficoll of PBMC. Cryopreserved PBMC taken from pediatric SOT recipients before 2019 used as non-infected controls (ctrls). PBMCs stimulated with 15-mer peptides from spike protein & co-stimulation with anti-CD49d & anti-CD28. Immunophenotyping by mass cytometry. Data analyzed using the Astrolabe Cytometry Platform, which follows a standardized pipeline for population identification & statistics & visualized using the t-SNE dimensionality reduction method. Disease severity stratified using WHO criteria. Subjects unvaccinated against COVID-19. Results: 38 children included: median age & time from transplant (TX) at infection: 5.5 (2.1–15.1) & 3.3 (0.2–14.7)-years for AI (n=10); 2.4 (0.6–15.6) & 0.5 (0.2–11.2)-years for ctrls (n=20); 6.7 (1.0–12.9) & 4.3 (0.2–9.7)-years for convalescence (n=8). NK & CD4 +T cell (EM, Th 2) frequency increased during AI {p<0.05}. CD57 expression low in T cells with EM phenotype, NKT, & gdT cells in AI vs. ctrls (p<0.05) & in NK cells in severe disease (p<0.05). CD161 expression high in CD4 +T cell (CM, Th1) in convalescence vs. ctrls{ p=0.04}. Conclusion: contraction in NK cells is not a feature of COVID-19 disease in pediatric SOT & global characteristics of NK, NKT & T cell compartments not compatible with senescence. Upregulation of cytolytic effectors in AI. Above a plausible mechanism for low mortality from COVID-19 disease in this patient population.

Detection and Prediction of Chipping in Wafer Grinding Based on Dicing Signal

Simple regression cannot wholly analyze large-scale wafer backside wall chipping because the wafer grinding process encounters many problems, such as collected data missing, data showing a non-linear distribution, and correlated hidden parameters lost. The objective of this study is to propose a novel approach to solving this problem. First, this study uses time series, random forest, importance analysis, and correlation analysis to analyze the signals of wafer grinding to screen out key grinding parameters. Then, we use PCA and Barnes-Hut t-SNE to reduce the dimensionality of the key grinding parameters and compare their corresponding heat maps to find out which dimensionality reduction method is more sensitive to the chipping phenomenon. Finally, this study imported the more sensitive dimensionality reduction data into the Data Driven-Bidirectional LSTM (DD-BLSTM) model for training and predicting the wafer chipping. It can adjust the key grinding parameters in time to reduce the occurrence of large-scale wafer chipping and can effectively improve the degree of deterioration of the grinding blade. As a result, the blades can initially grind three pieces of the wafers without replacement and successfully expand to more than eight pieces of the wafer. The accuracy of wafer chipping prediction using DD-BLSTM with Barnes-Hut t-SNE dimensionality reduction can achieve 93.14%.

Naïve pluripotent-like characteristics of non-tumorigenic Muse cells isolated from human amniotic membrane

Multilineage-differentiating stress-enduring (Muse) cells are non-tumorigenic pluripotent-like stem cells that exhibit triploblastic differentiation and self-renewability at the single-cell level, and are collectable as pluripotent surface marker SSEA-3(+) from the bone marrow (BM), peripheral blood, and organ connective tissues. SSEA-3(+) cells from human amniotic membrane mesenchymal stem cells (hAMSCs) were compared with hBM-Muse cells. Similar to hBM-Muse cells, hAMSC-SSEA-3(+) cells expressed pluripotency genes (OCT3/4, NANOG, and SOX2), differentiated into triploblastic cells from a single cell, self-renewed, and exhibited non-tumorigenicity. Notably, however, they exhibited unique characteristics not seen in hBM-Muse cells, including higher expression of genes related to germline- and extraembryonic cell-lineages compared with those in hBM-Muse cells in single-cell RNA-sequencing; and enhanced expression of markers relevant to germline- (PRDM14, TFAP2C, and NANOS3) and extraembryonic cell- (CDX2, GCM1, and ID2) lineages when induced by cytokine subsets, suggesting a broader differentiation potential similar to naïve pluripotent stem cells. t-SNE dimensionality reduction and Gene ontology analysis visualized hAMSC-SSEA-3(+) cells comprised a large undifferentiated subpopulation between epithelial- and mesenchymal-cell states and a small mesenchymal subpopulation expressing genes relevant to the placental formation. The AM is easily accessible by noninvasive approaches. These unique cells are a potentially interesting target naïve pluripotent stem cell-like resource without tumorigenicity.

A Malware Detection Scheme via Smart Memory Forensics for Windows Devices

With the introduction of 4G/5G Internet and the increase in the number of users, the malicious cyberattacks on computing devices have been increased making them vulnerable to external threats. High availability windows servers are designed to ensure delivery of consistent services such as business activities and e-services to their customers without any interruption. At the same time, a cyberattack on any of the clustered computer can put servers and customer devices in danger. A memory dump mechanism can capture the contents of memory in the event of a system or device crash such as corrupted files, damaged hardware, or irregular CPU power consumption. In this paper, we present a smart memory forensics scheme to recognize malicious attacks over high availability servers by capturing the memory dump of suspicious processes in the form of RGB visual images. Second, the local and global properties of malware images are captured using local binary patterns (LBP) and gray-level co-occurrence matrices (GLCM). A state-of-the-art t-distributed stochastic neighbor embedding scheme (t-SNE) is applied to reduce data dimensionality and improve the detection time of unknown malwares and their variants. An optimized CNN model is designed to predict malicious files harming servers or user devices. Throughout this study, we employed public data set of 4294 malicious samples covering malware variants and benign executables. A baseline is prepared to compare the performance of proposed model with state-of-the-art malware detection methods. The combined LBP + GLCM feature extraction along with t-SNE dimensionality reduction scheme further improved the detection accuracy by 98%, whereas the detection time is also increased by 73x. The overall performance shows that memory forensics is more effective for malware detection in terms accuracy and response time.

Exploring biomarkers and transcriptional factors in type 2 diabetes by comprehensive bioinformatics analysis on RNA-Seq and scRNA-Seq data.

BackgroundType 2 diabetes (T2D) is a prevalent chronic disease with elusive. Combining transcriptome and single-cell sequencing data to explore biomarkers of T2D could provide new insights into the in-depth understanding of the molecular mechanisms and diagnosis of T2D.MethodsThe GSE41762 dataset including RNA-seq data for healthy and T2D patients, was obtained from the Gene Expression Omnibus (GEO) database. The potential functions of the differentially expressed genes (DEGs) were revealed by Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis. Moreover, biomarkers were screened out by the Least Absolute Shrinkage and Selection Operator (LASSO) algorithm and receiver operating characteristic (ROC) analysis. Furthermore, single-cell RNA (sc-RNA)-seq data in the “E-MTAB-5061” dataset was downloaded from the ArrayExpress (European Bioinformatics Institute, EBI) database. Principal components analysis (PCA) and t-distributed stochastic neighbor embedding (tSNE) were used for dimensionality reduction analysis and cell clustering. The FindAllMarkers function was used annotate different cell clusters, and key cell clusters were screened by the expression levels of the biomarkers. Finally, the transcription factors (TFs) of the biomarkers were recognized.ResultsA total of 111 DEGs were screened in the GSE41762 dataset, which were mainly related to hormone secretion, specialized postsynaptic membrane, pancreatic secretion, JAK-STAT signaling pathway, and Ras signaling pathway. In addition, SLC2A2, SERPINF1, RASGRP1, and CHL1 were screened out as biomarkers of T2D, which possessed potential diagnostic value as AUC value greater than 0.8. A total of 1,515 T2D group cells and 1,817 healthy cohort cells were screened as core cells in the “E-MTAB-5061” dataset. Following tSNE dimensionality reduction cluster analysis, the core cells were divided into 13 cell clusters. According to the marker genes, the 13 cell clusters were annotated into six types of cells. Notably, SERPINF1 was highly expressed in fibroblasts and might be regulated by NR2F2 (nuclear receptor subfamily2, group F, and member 2).ConclusionsThis study identified four biomarkers (SLC2A2, SERPINF1, RASGRP1, and CHL1) for T2D, which provided new markers for the clinical diagnosis of T2D. Among them, SERPINF1 might be regulated by NR2F2, which provides valuable insight into the pathogensis of T2D.

Abstract 2708: Toward improved cancer classification using PCA + tSNE dimensionality reduction on bulk RNA-seq data

Abstract Intro: Minor variations in cancer type can have a major impact on therapeutic effectiveness and on the course of drug research and development. In order to improve upon existing -omic data classification methods, this study seeks to correctly classify previously unknown cancer samples and to determine -omic overlap between cancers by clustering patient RNA-seq data using PCA+tSNE for dimensionality reduction. Background: Although there have been major innovations in the use of bulk RNA-seq data for in silico modeling, RNA-seq analysis has limitations. RNA-seq data has over 20,000 dimensions, making it difficult to perform clustering and differential analysis important for in silico modeling. Thus, dimensionality reduction is a key for RNA-seq analysis. A novel method typically used in single-cell RNA-seq analysis combines PCA with tSNE. A common argument against the use of tSNE beyond data visualization is that the results are inconsistent and can vary depending on the hyperparameters of tSNE. This work demonstrates that combining PCA with tSNE creates a robust dimensional reduction of bulk RNA-seq data that allows for more accurate clustering of patient samples. Methods: Using the TCGA dataset with over 38 cancer subtypes and 10,351 total samples, the dimension was reduced to 50 principal components using PCA. TSNE was then applied with 1000-2000 iterations, a learning rate of 200, and varying perplexity of 5-50. Robustness was measured by how cancers of the same subtype are grouped in the lower dimension, in this case 2D. Agglomerative and k-means clustering algorithms were applied to the dimensional reduction results. This was done to track the accuracy of clustering on cancer subtypes and the movement of samples between these clusters. Results: For pure tSNE without PCA pre-reduction, clustering was not robust and showed less distinction between cancer subtypes and an increased number of outliers. Clustering accuracy was 50-60% for both k-means and agglomerative models and all varied hyperparameters. PCA+tSNE produced the most robust results with 60-70% accuracy for both k-means and agglomerative and all varied hyperparameters. It was observed that a low perplexity of 5-20, a learning rate of 100-300, and higher iterations (>1,000) produced the best results. Conclusions: PCA+tSNE is able to create accurate low dimensional representations of patient RNA-seq data that can be used to determine similarities and differences between patient samples based on gene expression data. The 10% variance in clustering accuracy suggests the method is robust and may help inform research and treatment by more robustly classifying cancer samples and by identifying similarities between cancers, especially among underserved, misdiagnosed, and rare cancers. Citation Format: Michael Bocker, Mikhail G. Grushko, Katherine E. Arline. Toward improved cancer classification using PCA + tSNE dimensionality reduction on bulk RNA-seq data [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 2708.