Nonzero Coefficients Research Articles

Bayesian quantile inference and order shrinkage for hysteretic quantile autoregressive models

Hysteretic quantile autoregressive model combines the hysteretic patterns and quantile autoregression, which can capture the dynamic relationship and nonlinear characteristics at different quantiles in time series data. In this paper, the Bayesian quantile inference and order shrinkage are studied for a class of hysteretic quantile autoregressive time series models. By using Markov Chain Monte Carlo (MCMC) techniques, the proposed Bayesian quantile method can handle the sparse hysteretic quantile autoregressive model well. It can accurately determine order of the model and estimate non-zero coefficients. Both simulation studies and a data example show that the proposed methods are feasible, reliable and appropriate for analysing the US Gross National Product data set.

Construction and validation of an endoscopic ultrasonography-based ultrasomics nomogram for differentiating pancreatic neuroendocrine tumors from pancreatic cancer

ObjectivesTo develop and validate various ultrasomics models based on endoscopic ultrasonography (EUS) for retrospective differentiating pancreatic neuroendocrine tumors (PNET) from pancreatic cancer.MethodsA total of 231 patients, comprising 127 with pancreatic cancer and 104 with PNET, were retrospectively enrolled. These patients were randomly divided into either a training or test cohort at a ratio of 7:3. Ultrasomics features were extracted from conventional EUS images, focusing on delineating the region of interest (ROI) for pancreatic lesions. Subsequently, dimensionality reduction of the ultrasomics features was performed by applying the Mann-Whitney test and least absolute shrinkage and selection operator (LASSO) algorithm. Eight machine learning algorithms, namely logistic regression (LR), light gradient boosting machine (LightGBM), multilayer perceptron (MLP), random forest (RF), extra trees, k nearest neighbors (KNN), support vector machine (SVM), and extreme gradient boosting (XGBoost), were employed to train prediction models using nonzero coefficient features. The optimal ultrasomics model was determined using a ROC curve and utilized for subsequent analysis. Clinical-ultrasonic features were assessed using both univariate and multivariate logistic regression. An ultrasomics nomogram model, integrating both ultrasomics and clinical-ultrasonic features, was developed.ResultsA total of 107 EUS-based ultrasomics features were extracted, and 6 features with nonzero coefficients were ultimately retained. Among the eight ultrasomics models based on machine learning algorithms, the RF model exhibited superior performance with an AUC= 0.999 (95% CI 0.9977 - 1.0000) in the training cohort and an AUC= 0.649 (95% CI 0.5215 - 0.7760) in the test cohort. A clinical-ultrasonic model was established and evaluated, yielding an AUC of 0.999 (95% CI 0.9961 - 1.0000) in the training cohort and 0.847 (95% CI 0.7543 - 0.9391) in the test cohort. Subsequently, the ultrasomics nomogram demonstrated a significant improvement in prediction accuracy in the test cohort, as evidenced by an AUC of 0.884 (95% CI 0.8047 - 0.9635) and confirmed by the Delong test. The calibration curve and decision curve analysis (DCA) depicted this ultrasomics nomogram demonstrated superior accuracy. They also yielded the highest net benefit for clinical decision-making compared to alternative models.ConclusionsA novel ultrasomics nomogram was proposed and validated, that integrated clinical-ultrasonic and ultrasomics features obtained through EUS, aiming to accurately and efficiently identify pancreatic cancer and PNET.

Mixed strategy approach destabilizes cooperation in finite populations with clustering coefficient.

Evolutionary game theory, encompassing discrete, continuous, and mixed strategies, is pivotal for understanding cooperation dynamics. Discrete strategies involve deterministic actions with a fixed probability of one, whereas continuous strategies employ intermediate probabilities to convey the extent of cooperation and emphasize expected payoffs. Mixed strategies, though akin to continuous ones, calculate immediate payoffs based on the action chosen at a given moment within intermediate probabilities. Although previous research has highlighted the distinct impacts of these strategic approaches on fostering cooperation, the reasons behind the differing levels of cooperation among these approaches have remained somewhat unclear. This study explores how these strategic approaches influence cooperation in the context of the prisoner's dilemma game, particularly in networked populations with varying clustering coefficients. Our research goes beyond existing studies by revealing that the differences in cooperation levels between these strategic approaches are not confined to finite populations; they also depend on the clustering coefficients of these populations. In populations with nonzero clustering coefficients, we observed varying degrees of stable cooperation for each strategic approach across multiple simulations, with mixed strategies showing the most variability, followed by continuous and discrete strategies. However, this variability in cooperation evolution decreased in populations with a clustering coefficient of zero, narrowing the differences in cooperation levels among the strategies. These findings suggest that in more realistic settings, the robustness of cooperation systems may be compromised, as the evolution of cooperation through mixed and continuous strategies introduces a degree of unpredictability.

Investigating dispersive optical solitons with the generalized stochastic perturbed Schrödinger–Hirota equation incorporating power-law nonlinearity and multiplicative white noise

This article aims to introduce the generalized stochastic perturbed Schrödinger–Hirota equation, which incorporates multiplicative white noise in the Itô sense. The study focuses on investigating the stochastic optical soliton solutions related to the governing model to understand the characteristics and properties of these soliton solutions. We can achieve this by utilizing the enhanced Kudryashov method and the enhanced direct algebraic approach. The study stands out for its innovative approach to examining the effects of multiplicative white noise on stochastic optical soliton solutions through the lens of the generalized stochastic perturbed Schrödinger–Hirota equation. Utilizing the enhanced Kudryashov method alongside the novel enhanced direct algebraic method, this research effectively identifies and delineates a range of soliton solution types, such as bright, dark, singular, and straddled solitons. This thorough investigation not only underscores the efficacy and efficiency of these methods but also sheds light on the impact of noise on the structural and dynamic properties of soliton solutions, supported by illustrative visualizations. A notable finding is that all solutions derived using the enhanced direct algebraic method feature a nonzero coefficient for the highest-order term of the auxiliary equation. The diversity of solutions obtained through these approaches includes bright, dark, singular, and straddled solitons. Additionally, the study reveals Weierstrass and Jacobi elliptic doubly periodic solutions, which can transition into various soliton types under specific parameter conditions, offering a broader understanding of soliton behavior and properties.

Representing polynomial of ST-CONNECTIVITY

We show that the coefficients of the representing polynomial of any monotone Boolean function are the values of the M\"obius function of an atomistic lattice related to this function. Using this we determine the representing polynomial of any Boolean function corresponding to a ST-CONNECTIVITY problem in acyclic quivers (directed acyclic multigraphs). Only monomials corresponding to unions of paths have non-zero coefficients which are $(-1)^D$ where $D$ is an easily computable function of the quiver corresponding to the monomial (it is the number of plane regions in the case of planar graphs). We determine that the number of monomials with non-zero coefficients for the two-dimensional $n \times n$ grid connectivity problem is $2^{\Omega(n^2)}$.

Multiomics Picture of Obesity in Young Adults.

Obesity is a socially significant disease that is characterized by a disproportionate accumulation of fat. It is also associated with chronic inflammation, cancer, diabetes, and other comorbidities. Investigating biomarkers and pathological processes linked to obesity is especially vital for young individuals, given their increased potential for lifestyle modifications. By comparing the genetic, proteomic, and metabolomic profiles of individuals categorized as underweight, normal, overweight, and obese, we aimed to determine which omics layer most accurately reflects the phenotypic changes in an organism that result from obesity. We profiled blood plasma samples by employing three omics methodologies. The untargeted GC×GC-MS metabolomics approach identified 313 metabolites. To augment the metabolomic dataset, we integrated a label-free HPLC-MS/MS proteomics method, leading to the identification of 708 proteins. The genomic layer encompassed the genotyping of 647,250 SNPs. Utilizing omics data, we trained sparse Partial Least Squares models to predict body mass index. Molecular features exhibiting frequently non-zero coefficients were selected as potential biomarkers, and we further explored enriched biological pathways. Proteomics was the most effective in single-omics analyses, with a median absolute error (MAE) of 5.44 ± 0.31 kg/m2, incorporating an average of 24 proteins per model. Metabolomics showed slightly lower performance (MAE = 6.06 ± 0.33 kg/m2), followed by genomics (MAE = 6.20 ± 0.34 kg/m2). As expected, multiomic models demonstrated better accuracy, particularly the combination of proteomics and metabolomics (MAE = 4.77 ± 0.33 kg/m2), while including genomics data did not enhance the results. This manuscript is the first multiomics study of obesity in a gender-balanced cohort of young adults profiled by genomic, proteomic, and metabolomic methods. The comprehensive approach provides novel insights into the molecular mechanisms of obesity, opening avenues for more targeted interventions.

Analysis of related factors for sarco-osteoporosis in middle-aged and elderly inpatients and development and validation of a nomogram

BackgroundSarco-osteoporosis is a skeletal muscle disease associated with aging and complex pathological factors. At present, there are few studies on the analysis of its related factors, and a nomogram to estimate the risk of sarco-osteoporosis in middle-aged and elderly patients is not available.MethodsA total of 386 patients admitted to our hospital from October 2021 to October 2022 were collected, and the general demographic data and clinical data of the patients were collected.386 subjects were enrolled in the study and randomly divided into training set and validation set at a ratio of 7:3. In the training set, the Least absolute shrinkage and selection operator(LASSO)regression technique was used to select the optimal predictive features, and multivariate logistic regression was used to screen the factors associated with sarco-osteoporosis, and a nomogram was constructed using meaningful variables from multivariate analysis. The performance of the nomograms was assessed and validated by Area Under Curve (AUC) and calibration curves.ResultsThere were no significant differences in baseline characteristic of individuals in training set and validation set, six variables with non-zero coefficients were screened based on LASSO regression in the training set. Multivariate logistic regression analysis showed that the related factors for sarco-osteoporosis in middle-aged and elderly inpatients included age (OR = 1.08, 95%CI 1.03 ∼ 1.14), regular exercise (OR = 0.29, 95%CI 0.15 ∼ 0.56), albumin (OR = 0.9, 95%CI 0.82 ∼ 0.98), height (OR = 0.93, 95%CI 0.88 ∼ 0.99) and lean mass index (OR = 0.66, 95%CI 0.52 ∼ 0.85), and a nomogram was constructed based on the above factors. AUC of nomogram were 0.868(95%CI 0.825 ∼ 0.912) in the training set and 0.737(95%CI 0.646 ∼ 0.828) in the validation set. Calibration curve analysis showed that the predicted probability of sarco-osteoporosis had high consistency with the actual probability, and the absolute error of the training set and verification set was 0.018 and 0.03, respectively.ConclusionsOur research showed that the occurrence of sarco-osteoporosis was associated with age, regular exercise, albumin, height and lean mass index, and we have developed a nomogram that can be effectively used in the preliminary and in-depth risk prediction of sarco-osteoporosis in middle-aged and elderly hospitalized patients.

Development and validation of a Radiopathomics model based on CT scans and whole slide images for discriminating between Stage I-II and Stage III gastric cancer

ObjectiveThis study aimed to develop and validate an artificial intelligence radiopathological model using preoperative CT scans and postoperative hematoxylin and eosin (HE) stained slides to predict the pathological staging of gastric cancer (stage I-II and stage III).MethodsThis study included a total of 202 gastric cancer patients with confirmed pathological staging (training cohort: n = 141; validation cohort: n = 61). Pathological histological features were extracted from HE slides, and pathological models were constructed using logistic regression (LR), support vector machine (SVM), and NaiveBayes. The optimal pathological model was selected through receiver operating characteristic (ROC) curve analysis. Machine learnin algorithms were employed to construct radiomic models and radiopathological models using the optimal pathological model. Model performance was evaluated using ROC curve analysis, and clinical utility was estimated using decision curve analysis (DCA).ResultsA total of 311 pathological histological features were extracted from the HE images, including 101 Term Frequency-Inverse Document Frequency (TF-IDF) features and 210 deep learning features. A pathological model was constructed using 19 selected pathological features through dimension reduction, with the SVM model demonstrating superior predictive performance (AUC, training cohort: 0.949; validation cohort: 0.777). Radiomic features were constructed using 6 selected features from 1834 radiomic features extracted from CT scans via SVM machine algorithm. Simultaneously, a radiopathomics model was built using 17 non-zero coefficient features obtained through dimension reduction from a total of 2145 features (combining both radiomics and pathomics features). The best discriminative ability was observed in the SVM_radiopathomics model (AUC, training cohort: 0.953; validation cohort: 0.851), and clinical decision curve analysis (DCA) demonstrated excellent clinical utility.ConclusionThe radiopathomics model, combining pathological and radiomic features, exhibited superior performance in distinguishing between stage I-II and stage III gastric cancer. This study is based on the prediction of pathological staging using pathological tissue slides from surgical specimens after gastric cancer curative surgery and preoperative CT images, highlighting the feasibility of conducting research on pathological staging using pathological slides and CT images.

On the security of JPEG image encryption with RS pairs permutation

Recently, a JPEG encryption scheme with RS (run/size) pairs multi-permutation and block permutation has been proposed, which mainly improves security by extracting the expected number of RS pairs for global permutations that need overflow processing. However, it still has the risk of information leakage under the chosen plaintext attack. This is because the limited embedding capacity in overflow processing makes the overflow blocks (OBs) less. On the one hand, a few OBs make the sketch generated by non-zero coefficients count after RS pair multi-permutation visually indistinguishable from the original sketch. On the other hand, a few OBs limit the number of extracted RS pairs that can participate in global permutation, so there are many partially extracted blocks (PEBs) containing unextracted RS pairs. More PEBs make the accuracy of the estimated block permutation sequence high, because the unique plaintext block can be constructed as long as there is an unextracted RS pair in an encrypted block. Simulation results show that the correct ratio of block permutation sequence is not less than 60 %. The highest and average PSNRs between the attacked sketch and the original one reach up to 47.07 dB and 27.42 dB, respectively.

High-dimensional nonconvex LASSO-type [formula omitted]-estimators

A theory is developed to examine the convergence properties of ℓ1-norm penalized high-dimensional M-estimators, with nonconvex risk and unrestricted domain. Under high-level conditions, the estimators are shown to attain the rate of convergence s0log(nd)/n, where s0 is the number of nonzero coefficients of the parameter of interest. Sufficient conditions for our main assumptions are then developed and finally used in several examples including robust linear regression, binary classification and nonlinear least squares.

JPEG-compatible Joint Image Compression and Encryption Algorithm with File Size Preservation

Joint image compression and encryption algorithms are intensively investigated due to their powerful capability of simultaneous image data compression and sensitive information protection. Unfortunately, most of the existing algorithms suffered from either poor compression efficiency or weak encryption strength, making them vulnerable to cryptanalysis. To address these limitations, we propose a chaos-based JPEG-compatible joint image compression and encryption algorithm. We separate the luminance and chrominance coefficients to preserve file size and encrypt the discrete cosine transform (DCT) coefficients in parallel. The proposed inter-block DC encryption strategy achieves high encryption intensity based on the permutation-substitution structure. In addition, we apply both inter- and intra-block permutations to AC coefficients and strengthen the encryption using an inter-block substitution for non-zero AC coefficients. The results of security and performance analyses demonstrate that the proposed algorithm offers robust encryption of image data while maintaining compression efficiency for real-time transmission.

Optimizer's dilemma: optimization strongly influences model selection in transcriptomic prediction.

Most models can be fit to data using various optimization approaches. While model choice is frequently reported in machine-learning-based research, optimizers are not often noted. We applied two different implementations of LASSO logistic regression implemented in Python's scikit-learn package, using two different optimization approaches (coordinate descent, implemented in the liblinear library, and stochastic gradient descent, or SGD), to predict mutation status and gene essentiality from gene expression across a variety of pan-cancer driver genes. For varying levels of regularization, we compared performance and model sparsity between optimizers. After model selection and tuning, we found that liblinear and SGD tended to perform comparably. liblinear models required more extensive tuning of regularization strength, performing best for high model sparsities (more nonzero coefficients), but did not require selection of a learning rate parameter. SGD models required tuning of the learning rate to perform well, but generally performed more robustly across different model sparsities as regularization strength decreased. Given these tradeoffs, we believe that the choice of optimizers should be clearly reported as a part of the model selection and validation process, to allow readers and reviewers to better understand the context in which results have been generated. The code used to carry out the analyses in this study is available at https://github.com/greenelab/pancancer-evaluation/tree/master/01_stratified_classification. Performance/regularization strength curves for all genes in the Vogelstein et al.(2013) dataset are available at https://doi.org/10.6084/m9.figshare.22728644.

Nash equilibria in risk-sensitive Markov stopping games under communication conditions

<p>This paper analyzes the existence of Nash equilibrium in a discrete-time Markov stopping game with two players. At each decision point, Player Ⅱ is faced with the choice of either ending the game and thus granting Player Ⅰ a final reward or letting the game continue. In the latter case, Player Ⅰ performs an action that affects transitions and receives a running reward from Player Ⅱ. We assume that Player Ⅰ has a constant and non-zero risk sensitivity coefficient, while Player Ⅱ strives to minimize the utility of Player Ⅰ. The effectiveness of decision strategies was measured by the risk-sensitive expected total reward of Player Ⅰ. Exploiting mild continuity-compactness conditions and communication-ergodicity properties, we found that the value function of the game is described as a single fixed point of the equilibrium operator, determining a Nash equilibrium. In addition, we provide an illustrative example in which our assumptions hold.</p>

Development of a nomogram to predict the risk of secondary failure of platelet recovery in patients with β-thalassemia major after hematopoietic stem cell transplantation: a retrospective study.

Secondary failure of platelet recovery (SFPR) is a common complication that influences survival and quality of life of patients with β-thalassemia major (β-TM) after hematopoietic stem cell transplantation (HSCT). A model to predict the risk of SFPR in β-TM patients after HSCT was developed. A retrospective study was used to develop the prediction model. The clinical data for 218 β-TM patients who received HSCT comprised the training set, and those for another 89 patients represented the validation set. The least absolute shrinkage and selection operator regression algorithm was used to identify the critical clinical factors with nonzero coefficients for constructing the nomogram. Calibration curve, C-index, and receiver operating characteristic curve assessments and decision curve analysis (DCA) were used to evaluate the calibration, discrimination, accuracy, and clinical usefulness of the nomogram. Internal and external validation were used to test and verify the predictive model. The nomogram based on pretransplant serum ferritin, hepatomegaly, mycophenolate mofetil use, and posttransplant serum albumin could be conveniently used to predict the SFPR risk of thalassemia patients after HSCT. The calibration curve of the nomogram revealed good concordance between the training and validation sets. The nomogram showed good discrimination with a C-index of 0.780 (95% CI: 70.3-85.7) and 0.868 (95% CI: 78.5-95.1) and AUCs of 0.780 and 0.868 in the training and validation sets, respectively. A high C-index value of 0.766 was reached in the interval validation assessment. DCA confirmed that the nomogram was clinically useful when intervention was decided at the possibility threshold ranging from 3% to 83%. We constructed a nomogram model to predict the risk of SFPR in patients with β-TM after HSCT. The nomogram has a good predictive ability and may be used by clinicians to identify SFPR patients early and recommend effective preventive measures.

Privacy Preserving Feature Selection for Sparse Linear Regression

Privacy-Preserving Machine Learning (PPML) provides protocols for learning and statistical analysis of data that may be distributed amongst multiple data owners (e.g., hospitals that own proprietary healthcare data), while preserving data privacy. The PPML literature includes protocols for various learning methods, including ridge regression. Ridge regression controls the L2 norm of the model, but does not aim to strictly reduce the number of non-zero coefficients, namely the L0 norm of the model. Reducing the number of non-zero coefficients (a form of feature selection) is important for avoiding overfitting, and for reducing the cost of using learnt models in practice. In this work, we develop a first privacy-preserving protocol for sparse linear regression under L0 constraints. The protocol addresses data contributed by several data owners (e.g., hospitals). Our protocol outsources the bulk of the computation to two non-colluding servers, using homomorphic encryption as a central tool. We provide a rigorous security proof for our protocol, where security is against semi-honest adversaries controlling any number of data owners and at most one server. We implemented our protocol, and evaluated performance with nearly a million samples and up to 40 features.

Correlation between general quality of life and oral health related quality in the mixed dentition.

The aim of this study was to evaluate the convergence between the domains of the Autoquestionnaire Qualité de Vie Enfant image (AUQUEI) and the Child Perceptions Questionnaire (CPQ8-10) in the mixed dentition. A sample of 676 children aged 8 to 10 years responded to the health-related quality of life (HRQoL) and oral health-related quality of life (OHRQoL) questionnaires using the AUQUEI and the CPQ8-10, respectively. Clinical (dental caries and malocclusion) and socioeconomic variables were assessed. The validity of convergence between scores (total and per domain) of the two instruments was assessed by Spearman correlation analysis, considering that non-zero coefficient values represented a correlation between scores. The median was calculated to compare the scores of each questionnaire relative to the variables, and the nonparametric Mann-Whitney test was applied to determine statistically significant differences between the categories. A weak significant correlation (between 0.30 and 0.50) was observed between the domains and the total scores of instruments (p < 0.05), except for the leisure domain (p > 0.05). Participants with a lower family income had worse HRQoL (p < 0.05), and those with caries and malocclusion experience had worse OHRQoL (p < 0.05). In conclusion, the AUQUEI and CPQ8-10 instruments showed a weak correlation. Income and clinical variables had a negative impact on the AUQUEI and CPQ8-10, respectively.

Nonlocal effective medium theory for phononic temporal metamaterials

We have developed a nonlocal effective medium theory (EMT) for phononic temporal metamaterials using the multiscale technique. Our EMT yields closed-form expressions for effective constitutive parameters and reveals these materials as reciprocal media with symmetric band dispersion. Even without spatial symmetry breaking, nonzero Willis coupling coefficients emerge with time modulation and broken time-reversal symmetry, when the nonlocal effect is taken into account. Compared to the local EMT, our nonlocal version is more accurate for calculating the bulk band at high wavenumbers and essential for understanding nonlocal effects at temporal boundaries. This nonlocal EMT can be a valuable tool for studying and designing phononic temporal metamaterials beyond the long-wavelength limit.

Association between Low Bone Mineral Density and Pain in a Prospective Cohort of Adults with Sickle Cell Disease: Preliminary Results from the SCD Bone Pain Study

Background: Osteonecrosis (ON), a prevalent skeletal complication of sickle cell disease (SCD), typically associates with severe chronic pain, frequent acute care utilization, and permanent disability. We previously showed that low bone mineral density (BMD) independently associates with hip ON and chronic pain in a pediatric SCD cohort. We now present preliminary data on the association between low BMD and pain in a prospective cohort of adults enrolled in the SCD Bone Pain Study (NCT05283148). Methods: Written informed consent was obtained from all participants, who were recruited from ambulatory adults with SCD seen at the University of California Davis Comprehensive Cancer Center. All participants underwent dual-energy X-ray absorptiometry (DXA) scans of their lumbar spine, left hip, and left forearm. Vertebral fractures were assessed by thoracolumbar morphometry in the DXA scanner. All participants also completed the Adult Sickle Cell Quality of Life Measurement Information System (ASCQ-Me) Pain Impact questionnaire. We used linear regression to evaluate the association between ASCQ-Me pain scores and low BMD (categorically defined as BMD Z-score ≤ -2 at any DXA site). We also used linear regression to determine the association between ASCQ-Me pain scores and continuous total hip and lumbar spine BMD Z-scores. Lastly, we evaluated the impact of other covariates on pain scores, with low BMD or BMD Z-scores included in the least absolute shrinkage and selection operator (LASSO, no shrinkage). Results: Thirty adults (median age 34.5 years [20-67 years]) completed the study between Nov 2022-Jul 2023. Baseline characteristics are summarized in Table 1. There was a borderline significant association between low BMD and lower (worse) pain scores (p=0.077). Continuous BMD Z-scores of the total hip (p=0.081) and lumbar spine (p=0.024) were independently associated with higher (better) pain scores. In addition to low BMD, only white blood cell count (WBC), indirect bilirubin, and vitamin D were identified as possible covariates (all had non-zero coefficients in LASSO analysis). Low BMD was significantly associated with pain (p=0.024), even after adjusting for WBC, indirect bilirubin, and vitamin D (Table 2). Similarly, we found that higher hip (p=0.059) and lumbar (p=0.027) BMD Z-scores were associated with higher (better) pain scores, despite adjusting for the same confounders. Discussion: We found low BMD in 12 of 30 (40%) adults enrolled in the SCD Bone Pain Study. Like other chronic hemolytic anemias, compensatory increased erythropoiesis leads to trabeculae degradation, cortical bone thinning, and eventual low bone mass. Lumbar vertebrae comprises mostly of trabecular bone, while the hip joint has mostly cortical bone. This might explain why mean lumbar BMD Z-scores were lower than hip BMD Z-scores (Table 1), and why lumbar spine BMD Z-scores directly and significantly correlated with pain scores (Table 2). These preliminary data suggest that pain is more prevalent in SCD adults with low BMD than those with normal BMD as we previously reported in a pediatric SCD cohort. Pain scores decrease (worsen) with increasing WBC, which suggests an inflammatory component to chronic SCD pain. Contrary to expectation, higher indirect bilirubin (a marker of increased hemolysis) significantly correlated with higher (better) pain scores. Similarly, higher vitamin D (a proxy for bone health) correlated with lower (worse) pain scores, though this association was non-significant. The reason(s) for these counterintuitive associations between pain scores, indirect bilirubin, and vitamin D levels is currently unknown. Limitations include a small sample size, though using the LASSO method for multivariate regression adjusted for this. Further, low BMD causality could not be determined at this time. Recruitment is ongoing with an enrollment target of 50 adults with SCD by Nov 2023. Analysis are ongoing to determine if SCD-modifying therapies will impact the association between low BMD and pain. Bone biomarker studies are also pending. Further, the thoracolumbar morphometry is still undergoing radiologic review, so we can further assess how vertebral compression fractures (i) correlate with pain scores, and/or (ii) attenuate the association between low BMD and pain. Identifying and treating risk factors for low BMD may alleviate bone pain and potentially improve quality of life for all people with SCD.

A DNA Methylation Signature Predicts Clinical Outcome in Pediatric AML Patients

Pediatric Acute Myeloid Leukemia (AML) is a heterogenous disease with a dismal outcome. Recently, transcriptomic-based leukemic stemness scores have highlighted the importance of DNA methyltransferase 3B in pediatric AML prognosis, which invites a deeper investigation of leukemic stem cells through the scope of their DNA methylome. Other studies have shown the prognostic impact of genome-wide methylation burden (PMID: 29928480, 33080932). As the World Health Organization has recognized methylation analysis as the most apt technology for determining lineage and cell population origins of tumors (PMID: 34921008), it is befitting to evaluate the DNA methylome as a prognostic tool in hematopoietic malignancies. Thus, we hypothesize that a DNA methylation-based signature has the potential to enhance the prognostic precision in pediatric AML. In this study, we applied machine learning approaches to DNA methylation data from 924 pediatric AML patients (only including data from samples with diagnostic bone marrow specimens) from three Children's Oncology Group clinical trials: AAML1031, AAML0531, and AAML03P1, available through GSE190931, GSE124413 and GDC_TARGET-AML, respectively. Clinical endpoints available were defined as :i) Minimal residual disease after induction 1 (MRD1): positive MRD1 if ≥ 1 leukemic cell per 1,000 mononuclear bone marrow cells (≥ 0.1%) determined by flow cytometry; ii) event-free survival (EFS) defined as the time from study enrollment to induction failure, relapse, secondary malignancy, death, or study withdrawal for any reason, with event-free patients censored on last follow-up; iii) overall survival (OS) defined as the time from study enrollment to death, with living patients censored on the date of last follow-up. Processing of the raw data followed best practices in the literature using SeSAMe (PMID: 30085201, 27924034). To independently validate the findings derived from the discovery cohort, we processed in parallel meDNA array data from bone marrow specimens at diagnosis from 201 AML patients treated on the multi-site clinical trials AML02 and AML08 (NCT00136084 and NCT00703820). As step 1, we performed Epigenome-Wide Association Study with Cox Proportional Hazards Regression (Cox-PH EWAS) to identify CpGs with methylation levels most predictive of OS, controlling for risk group categories. At p-value threshold of 1.0e-6, 167 CpGs were associated with OS, which were chosen as candidate CpGs for step 2 analysis using LASSO-based Cox-PH model, with 1,000 iterations of 10-fold cross-validation. 35 CpGs were consistently selected as non-zero coefficients in at least 95% of the models and were used to define the CpG based signature. Patients were further categorized into either Low MethylScoreAML Px (n=563, 60%) or High MethylScoreAML Px (n=361, 40%) groups using recursive partitioning statistics. In the discovery cohort, patients with High MethylScoreAML Px demonstrated a markedly poorer probability of OS (HR 3.91, p&amp;lt;0.0001) and EFS (HR 2.61, &amp;lt;0.0001) in comparison to the low MethylScore. These results were further tested in the validation cohort (n=201) for OS (HR 2.45, p=0.0006) and EFS (HR 2.15, p=0.0004). Notably, in subgroup analyses for both discovery and validation, MethylScoreAML Px remained statistically significant in multivariable models for EFS and OS with adjustments for MRD1 status, risk group, FLT3 status, white blood cell count at diagnosis, and patient age, indicating it as a robust, independent model supplementing currently known prognostic factors. In conclusion, this study unveils MethylScoreAML Px, a methylation-based prognostic risk score that robustly predicts clinical outcome in pediatric AML patients. To our knowledge, we are the first group to describe a Cox-PH-EWAS analysis in pediatric AML, as well as the first to devise and validate a prognostic risk score in &amp;gt;1000 patients from multiple pediatric-only trials. Future directions of research are focused on testing MethylScore in fresh peripheral blood samples using long read, real-time Oxford Nanopore sequencing technology, and comparing it with other risk scores.

Robust hybrid algorithms for regularization and variable selection in QSAR studies

This study introduces a robust hybrid sparse learning approach for regularization and variable selection. This approach comprises two distinct steps. In the initial step, we segment the original dataset into separate training and test sets and standardize the training data using its mean and standard deviation. We then employ either the LASSO or sparse LTS algorithm to analyze the training set, facilitating the selection of variables with non-zero coefficients as essential features for the new dataset. Secondly, the new dataset is divided into training and test sets. The training set is further divided into k folds and evaluated using a combination of Random Forest, Ridge, Lasso, and Support Vector Regression machine learning algorithms. We introduce novel hybrid methods and juxtapose their performance against existing techniques. To validate the efficacy of our proposed methods, we conduct a comprehensive simulation study and apply them to a real-life QSAR analysis. The findings unequivocally demonstrate the superior performance of our proposed estimator, with particular distinction accorded to SLTS+LASSO. In summary, the twostep robust hybrid sparse learning approach offers an effective regularization and variable selection applicable to a wide spectrum of real-world problems.