Self-consistency Test Research Articles

DeepDBS: Identification of DNA-binding sites in protein sequences by using deep representations and random forest

Interactions of biological molecules in organisms are considered to be primary factors for the lifecycle of that organism. Various important biological functions are dependent on such interactions and among different kinds of interactions, the protein DNA interactions are very important for the processes of transcription, regulation of gene expression, DNA repairing and packaging. Thus, keeping the knowledge of such interactions and the sites of those interactions is necessary to study the mechanism of various biological processes. As experimental identification through biological assays is quite resource-demanding, costly and error-prone, scientists opt for the computational methods for efficient and accurate identification of such DNA-protein interaction sites. Thus, herein, we propose a novel and accurate method namely DeepDBS for the identification of DNA-binding sites in proteins, using primary amino acid sequences of proteins under study. From protein sequences, deep representations were computed through a one-dimensional convolution neural network (1D-CNN), recurrent neural network (RNN) and long short-term memory (LSTM) network and were further used to train a Random Forest classifier. Random Forest with LSTM-based features outperformed the other models, as well as the existing state-of-the-art methods with an accuracy score of 0.99 for self-consistency test, 10-fold cross-validation, 5-fold cross-validation, and jackknife validation while 0.92 for independent dataset testing. It is concluded based on results that the DeepDBS can help accurate and efficient identification of DNA binding sites (DBS) in proteins.

DEL-Thyroid: deep ensemble learning framework for detection of thyroid cancer progression through genomic mutation

Genes, expressed as sequences of nucleotides, are susceptible to mutations, some of which can lead to cancer. Machine learning and deep learning methods have emerged as vital tools in identifying mutations associated with cancer. Thyroid cancer ranks as the 5th most prevalent cancer in the USA, with thousands diagnosed annually. This paper presents an ensemble learning model leveraging deep learning techniques such as Long Short-Term Memory (LSTM), Gated Recurrent Units (GRUs), and Bi-directional LSTM (Bi-LSTM) to detect thyroid cancer mutations early. The model is trained on a dataset sourced from asia.ensembl.org and IntOGen.org, consisting of 633 samples with 969 mutations across 41 genes, collected from individuals of various demographics. Feature extraction encompasses techniques including Hahn moments, central moments, raw moments, and various matrix-based methods. Evaluation employs three testing methods: self-consistency test (SCT), independent set test (IST), and 10-fold cross-validation test (10-FCVT). The proposed ensemble learning model demonstrates promising performance, achieving 96% accuracy in the independent set test (IST). Statistical measures such as training accuracy, testing accuracy, recall, sensitivity, specificity, Mathew's Correlation Coefficient (MCC), loss, training accuracy, F1 Score, and Cohen's kappa are utilized for comprehensive evaluation.

Bayesian F -statistic-based parameter estimation of continuous gravitational waves from known pulsars

We present a new method and implementation to obtain Bayesian posteriors on the amplitude parameters {h0,cosι,ψ,ϕ0} of continuous gravitational waves emitted by known pulsars. This approach leverages the well-established F-statistic framework and software. We further explore the benefits of employing a likelihood function that is analytically marginalized over ϕ0, which avoids signal degeneracy problems in the ψ−ϕ0 subspace. The method is tested on simulated signals, hardware injections in Advanced-LIGO detector data, and by performing percentile-percentile self-consistency tests of the posteriors via Monte-Carlo simulations. We apply our methodology to PSR J1526-2744, a recently discovered millisecond pulsar. We find no evidence for a signal and obtain a Bayesian upper limit h095% on the gravitational-wave amplitude of approximately 7×10−27, comparable with a previous frequentist upper limit. Published by the American Physical Society 2024

Antifp SRF: Identifying Antifungal Peptides by Sequence Statistical Moments and Random Forest Classifier

Numerous types of fungus are unhealthy and can lead to life-threatening conditions in people. A fungal infection that targets specific bodily regions and that the immune system is unable to fight off. Numerous studies have been conducted on the topic, but due to its negative side effects, medications, chemicals, and conventional techniques have not proven effective enough. Due to the rapidly invading fungus species, antimicrobial peptides having antifungal properties emerged as a new and powerful candidate due to their efficiency and selectivity. Many mathematical models are being developed for its identification, including Deep-AntiFP and AntiFP, the most recently. However, it has limited performance in accuracy, sensitivity, precision, and MCC. This study investigates an efficient and accurate computational model for the prediction of antifungal peptides (AFPs) using machine learning approaches. The model uses the statistical moment as feature extraction and Random Forest (RF) as a predictor to predict AFPs from non-AFPs. 10-fold cross-validation, independent tests, and self-consistency tests are used to validate the performance of the model. Three datasets are used for model training and its evaluation, namely Antifp Main, Antifp DS1, and Independent Main. The model achieved an accuracy of 95.73%, 97.01%, and 99.15% for Antifp Main, Antifp DS1, and Independent main respectively through independent testing; 97.4%, 97.3%, and 97.4% by 10-fold cross-validation and 99.74%, 99.87 and 99.83 by self-consistency testing. The proposed model has shown excellent results for all three datasets as compared to the existing models available in the literature.

Harmonizing pesticides environmental quality standards: A fate-pathway perspective

Regulatory agencies worldwide set pesticide environmental quality standards, which are proposed independently in each dependent environmental media rather than across the complete fate route. Thus, lacking the fate-pathway perspective in defining pesticide environmental quality standards might cause undesirable pesticide residue from the upper compartment (e.g., soil) to the lower compartment (e.g., water). This study aimed to harmonize the self-consistency of pesticide environmental quality standards across environmental media via the fate-pathway analysis. The introduced qualitative and quantitative rules defined environmental quality standards of pesticides in six major environmental scenarios in the soil and water system based on related regulatory objectives. Fate factors simulated via USEtox were used to create a preliminary quantitative link between theoretical maximum legal masses of pesticides across environmental compartments. Using chlorpyrifos and 2,4-D as examples, their standard values were comparatively assessed in selected environmental media in China and the United States. According to the investigative findings, missing the respective environmental quality standards of pesticides in the agricultural soil could significantly influence the implementation of those in freshwater. Taking a fate-pathway perspective, the self-consistency test highlighted that defining pesticide environmental quality standards for freshwater was the most challenging task, as the freshwater compartment typically comprises multiple lower environmental compartments with diverse regulatory objectives. Overall, this theoretical study has the potential to illuminate the harmonization of pesticide environmental quality standards throughout the entire environmental fate pathway, ultimately leading to improved regulatory efficiency and communication. Future research should focus on risk-based model implementation, regulatory response evaluation, and legal limit interpretation to better integrate environmental pesticide management under a variety of regulatory goals.

An ensemble-based deep learning model for detection of mutation causing cutaneous melanoma

When the mutation affects the melanocytes of the body, a condition called melanoma results which is one of the deadliest skin cancers. Early detection of cutaneous melanoma is vital for raising the chances of survival. Melanoma can be due to inherited defective genes or due to environmental factors such as excessive sun exposure. The accuracy of the state-of-the-art computer-aided diagnosis systems is unsatisfactory. Moreover, the major drawback of medical imaging is the shortage of labeled data. Generalized classifiers are required to diagnose melanoma to avoid overfitting the dataset. To address these issues, blending ensemble-based deep learning (BEDLM-CMS) model is proposed to detect mutation of cutaneous melanoma by integrating long short-term memory (LSTM), Bi-directional LSTM (BLSTM) and gated recurrent unit (GRU) architectures. The dataset used in the proposed study contains 2608 human samples and 6778 mutations in total along with 75 types of genes. The most prominent genes that function as biomarkers for early diagnosis and prognosis are utilized. Multiple extraction techniques are used in this study to extract the most-prominent features. Afterwards, we applied different DL models optimized through grid search technique to diagnose melanoma. The validity of the results is confirmed using several techniques, including tenfold cross validation (10-FCVT), independent set (IST), and self-consistency (SCT). For validation of the results multiple metrics are used which include accuracy, specificity, sensitivity, and Matthews’s correlation coefficient. BEDLM gives the highest accuracy of 97% in the independent set test whereas in self-consistency test and tenfold cross validation test it gives 94% and 93% accuracy, respectively. Accuracy of in self-consistency test, independent set test, and tenfold cross validation test is LSTM (96%, 94%, 92%), GRU (93%, 94%, 91%), and BLSTM (99%, 98%, 93%), respectively. The findings demonstrate that the proposed BEDLM-CMS can be used effectively applied for early diagnosis and treatment efficacy evaluation of cutaneous melanoma.

Schwinger poles of the three-gluon vertex: symmetry and dynamics

The implementation of the Schwinger mechanism endows gluons with a nonperturbative mass through the formation of special massless poles in the fundamental QCD vertices; due to their longitudinal character, these poles do not cause divergences in on-shell amplitudes, but induce detectable effects in the Green’s functions of the theory. Particularly important in this theoretical setup is the three-gluon vertex, whose pole content extends beyond the minimal structure required for the generation of a gluon mass. In the present work we analyze these additional pole patterns by means of two distinct, but ultimately equivalent, methods: the Slavnov–Taylor identity satisfied by the three-gluon vertex, and the nonlinear Schwinger–Dyson equation that governs the dynamical evolution of this vertex. Our analysis reveals that the Slavnov–Taylor identity imposes strict model-independent constraints on the associated residues, preventing them from vanishing. Approximate versions of these constraints are subsequently recovered from the Schwinger–Dyson equation, once the elements responsible for the activation of the Schwinger mechanism have been duly incorporated. The excellent coincidence between the two approaches exposes a profound connection between symmetry and dynamics, and serves as a nontrivial self-consistency test of this particular mass generating scenario.

Experimental solubility of aripiprazole in supercritical carbon dioxide and modeling

The solubility of compounds in supercritical carbon dioxide (SC-{mathrm{CO}}_{2}) has found crucial significance in the fabrication of micro/nano-scaled drugs. In this research, the solubility of Aripiprazole was measured in SC-{mathrm{CO}}_{2} at various temperatures (308–338 K) and pressures (12–30 MPa). Moreover, the experimental solubility results were correlated with several semi-empirical models (Chrastil, Bartle et al., Kumar & Johnston, Menden-Santiago & Teja, Sodeifian et al., and Jouyban et al.) as well as the modified Wilson model. The molar fraction of the drug in SC-{mathrm{CO}}_{2} varied in the range of 1.830times {10}^{-6} to 1.036times {10}^{-5}. The solubility highly depended on the operating pressure and temperature. The Chrastil (0.994), Jouyban et al. (0.993) and Sodeifian et al. (0.992) models showed the highest consistency with the obtained values. Furthermore, self-consistency tests were performed on the solubility of Aripiprazole in SC-{mathrm{CO}}_{2}. The approximate total enthalpy ({mathrm{Delta H}}_{mathrm{total}}), vaporization enthalpy ({mathrm{Delta H}}_{mathrm{vap}}), and solubility enthalpy ({mathrm{Delta H}}_{mathrm{sol}}) were also calculated.

EDLM: Ensemble Deep Learning Model to Detect Mutation for the Early Detection of Cholangiocarcinoma.

The most common cause of mortality and disability globally right now is cholangiocarcinoma, one of the worst forms of cancer that may affect people. When cholangiocarcinoma develops, the DNA of the bile duct cells is altered. Cholangiocarcinoma claims the lives of about 7000 individuals annually. Women pass away less often than men. Asians have the greatest fatality rate. Following Whites (20%) and Asians (22%), African Americans (45%) saw the greatest increase in cholangiocarcinoma mortality between 2021 and 2022. For instance, 60-70% of cholangiocarcinoma patients have local infiltration or distant metastases, which makes them unable to receive a curative surgical procedure. Across the board, the median survival time is less than a year. Many researchers work hard to detect cholangiocarcinoma, but this is after the appearance of symptoms, which is late detection. If cholangiocarcinoma progression is detected at an earlier stage, then it will help doctors and patients in treatment. Therefore, an ensemble deep learning model (EDLM), which consists of three deep learning algorithms-long short-term model (LSTM), gated recurrent units (GRUs), and bi-directional LSTM (BLSTM)-is developed for the early identification of cholangiocarcinoma. Several tests are presented, such as a 10-fold cross-validation test (10-FCVT), an independent set test (IST), and a self-consistency test (SCT). Several statistical techniques are used to evaluate the proposed model, such as accuracy (Acc), sensitivity (Sn), specificity (Sp), and Matthew's correlation coefficient (MCC). There are 672 mutations in 45 distinct cholangiocarcinoma genes among the 516 human samples included in the proposed study. The IST has the highest Acc at 98%, outperforming all other validation approaches.

The Hubble constant from galaxy cluster scaling-relation and SNe Ia observations: a consistency test

In this paper, we propose a self-consistent test for a Hubble constant estimate using galaxy cluster and type Ia supernovae (SNe Ia) observations. The approach consists, in a first step, of obtaining the observational value of the galaxy cluster scaling-relation Y_{SZE}D_{A}^{2}/C_{XSZ}Y_X = C by combining the X-Ray and SZ observations of galaxy clusters at low redshifts (z < 0.1) from the first Planck mission all-sky data set (0.044 le z le 0.444), along with SNe Ia observations and making use of the cosmic distance duality relation validity. Then, by considering a flat Lambda CDM model for D_A, the constant C from the first step and the Planck prior on Omega _M parameter, we obtain H_0 by using the galaxy cluster data with z>0.1. As a result, we obtain { H_0=73.014^{+7.435}_{-6.688}} km/s/Mpc, in full agreement with the latest results from HST + SH0ES team. We also compare our method with that one where the C parameter is obtained from hydrodynamical simulations of massive galaxy clusters.

Observational constraints of an anisotropic boost due to the projection effects using redMaPPer clusters

ABSTRACT Optical clusters identified from red-sequence galaxies suffer from projection effects, where interloper galaxies along the line of sight to a cluster are mistaken as genuine members of the cluster. In the previous study, we found that the projection effects cause the boost on the amplitudes of clustering and lensing on large scale compared to the expected amplitudes in the absence of any projection effects. These boosts are caused by preferential selections of filamentary structure aligned to the line of sight due to distance uncertainties in photometric surveys. We model the projection effects with two simple assumptions and develop a novel method to quantify the size of the boost using cluster-galaxy cross-correlation functions. We validate our method using mock cluster catalogues built from cosmological N-body simulations and find that we can obtain unbiased constraints on the boost parameter with our model. We then apply our analysis on the SDSS redMaPPer clusters and find that the size of the boost is roughly 20 per cent for all the richness bins except the cluster sample with the richness bin λ ∈ [30, 40]. This is the first study to constrain the boost parameter independent from cluster cosmology studies and provides a self-consistency test for the projection effects.

Solubility of Cortisone and Hydrocortisone in Supercritical Carbon Dioxide and Ethanol

Cortisone and hydrocortisone are poorly water-soluble corticosteroids widely used to treat many inflammatory and immune diseases. In the prospect of employing them in pharmaceutical applications or analytical techniques involving cosolvent-modified supercritical fluids, their solubility has been determined, for the first time, in a mixture of supercritical CO2 and 2, 3, or 4 mol % of ethanol, at 318.15, 328.15, 348.15, and 373.15 K, and in the pressure range of 13–27 MPa. Tests were conducted in a semi-flow laboratory apparatus, and results were fitted with the most popular density-based correlations reported in the literature for ternary systems. Even though the two drugs exhibit similar molecular structures, the solubility of cortisone in the ethanol-scCO2 mixtures is on average 3.5 times higher than that of hydrocortisone. This is congruent with the fact that the solubility of cortisone in pure ethanol is higher. The seven-parameter Reddy–Madras model (adjusted absolute average percent deviation (AARD%) = 8.8%) proved to be the best correlating one for cortisone while the six-parameter Garlapati–Madras approach (AARD% = 11.6%) provided the best fitting for hydrocortisone. The reliability of the experimental results was also confirmed by the positive response to the Méndez-Santiago–Teja self-consistency test.

Hemolytic-Pred: A machine learning-based predictor for hemolytic proteins using position and composition-based features.

The objective of this study is to propose a novel in-silico method called Hemolytic-Pred for identifying hemolytic proteins based on their sequences, using statistical moment-based features, along with position-relative and frequency-relative information. Primary sequences were transformed into feature vectors using statistical and position-relative moment-based features. Varying machine learning algorithms were employed for classification. Computational models were rigorously evaluated using four different validation. The Hemolytic-Pred webserver is available for further analysis at http://ec2-54-160-229-10.compute-1.amazonaws.com/. XGBoost outperformed the other six classifiers with an accuracy value of 0.99, 0.98, 0.97, and 0.98 for self-consistency test, 10-fold cross-validation, Jackknife test, and independent set test, respectively. The proposed method with the XGBoost classifier is a workable and robust solution for predicting hemolytic proteins efficiently and accurately. The proposed method of Hemolytic-Pred with XGBoost classifier is a reliable tool for the timely identification of hemolytic cells and diagnosis of various related severe disorders. The application of Hemolytic-Pred can yield profound benefits in the medical field.

Deep Learning Approaches for Detection of Breast Adenocarcinoma Causing Carcinogenic Mutations.

Genes are composed of DNA and each gene has a specific sequence. Recombination or replication within the gene base ends in a permanent change in the nucleotide collection in a DNA called mutation and some mutations can lead to cancer. Breast adenocarcinoma starts in secretary cells. Breast adenocarcinoma is the most common of all cancers that occur in women. According to a survey within the United States of America, there are more than 282,000 breast adenocarcinoma patients registered each 12 months, and most of them are women. Recognition of cancer in its early stages saves many lives. A proposed framework is developed for the early detection of breast adenocarcinoma using an ensemble learning technique with multiple deep learning algorithms, specifically: Long Short-Term Memory (LSTM), Gated Recurrent Units (GRU), and Bi-directional LSTM. There are 99 types of driver genes involved in breast adenocarcinoma. This study uses a dataset of 4127 samples including men and women taken from more than 12 cohorts of cancer detection institutes. The dataset encompasses a total of 6170 mutations that occur in 99 genes. On these gene sequences, different algorithms are applied for feature extraction. Three types of testing techniques including independent set testing, self-consistency testing, and a 10-fold cross-validation test is applied to validate and test the learning approaches. Subsequently, multiple deep learning approaches such as LSTM, GRU, and bi-directional LSTM algorithms are applied. Several evaluation metrics are enumerated for the validation of results including accuracy, sensitivity, specificity, Mathew’s correlation coefficient, area under the curve, training loss, precision, recall, F1 score, and Cohen’s kappa while the values obtained are 99.57, 99.50, 99.63, 0.99, 1.0, 0.2027, 99.57, 99.57, 99.57, and 99.14 respectively.

Supercritical carbon dioxide utilization in drug delivery: Experimental study and modeling of paracetamol solubility

In the present study, the solubility of paracetamol in supercritical CO2 is measured at temperatures between 311 and 358 K and pressures between 95 and 265 bar. It was shown that the solubility of paracetamol through a static solubility measurement method was between 0.3055 × 10−6 to 16.3582 × 10−6 based on mole fraction. The obtained experimental solubility data revealed the direct effect of pressure on the paracetamol experimental data, while the temperature has a dual effect of both increasing and decreasing effect considering the shifting point known as crossover pressure which was measured to be around 110 bar for paracetamol. Besides, two theoretical approaches were applied to predict the paracetamol experimental results. The experimental data were modeled with two various equations of state (EoSs) i.e. Peng-Robinson EoS and the simplified perturbed chain statistical associating fluid theory (sPC-SAFT) EoS, which their binary interaction parameters were optimized by using genetic algorithm. The obtained results demonstrated the sPC-SAFT EoS predicted the solubility of paracetamol with more accuracy (the average percent deviation was equal to 2.5215), especially at higher pressures. Furthermore, four well-known semi-empirical density-based correlations namely Mendez-Santiago and Teja (MST), Kumar-Johnston (KJ), Bartle et al., and Garlapati and Madras models were applied for modeling the solubility of paracetamol experimental data. According to self-consistency test, KJ model presented more accurate correlative capability with average percent deviation about 4.09%. As a final point, the rapid expansion of supercritical solution process was performed to reduce particle size of paracetamol and mean particle size of paracetamol was calculated based on EoSs and mathematical modeling.

Machine learning techniques for identification of carcinogenic mutations, which cause breast adenocarcinoma

Breast adenocarcinoma is the most common of all cancers that occur in women. According to the United States of America survey, more than 282,000 breast cancer patients are registered each year; most of them are women. Detection of cancer at its early stage saves many lives. Each cell contains the genetic code in the form of gene sequences. Changes in the gene sequences may lead to cancer. Replication and/or recombination in the gene base sometimes lead to a permanent change in the nucleotide sequence of the genome, called a mutation. Cancer driver mutations can lead to cancer. The proposed study develops a framework for the early detection of breast adenocarcinoma using machine learning techniques. Every gene has a specific sequence of nucleotides. A total of 99 genes are identified in various studies whose mutations can lead to breast adenocarcinoma. This study uses the dataset taken from 4127 human samples, including men and women from more than 12 cohorts. A total of 6170 mutations in gene sequences are used in this study. Decision Tree, Random Forest, and Gaussian Naïve Bayes are applied to these gene sequences using three evaluation methods: independent set testing, self-consistency testing, and tenfold cross-validation testing. Evaluation metrics such as accuracy, specificity, sensitivity, and Mathew’s correlation coefficient are calculated. The decision tree algorithm obtains the best accuracy of 99% for each evaluation method.

IAcety-SmRF: Identification of Acetylation Protein by Using Statistical Moments and Random Forest.

Acetylation is the most important post-translation modification (PTM) in eukaryotes; it has manifold effects on the level of protein that transform an acetyl group from an acetyl coenzyme to a specific site on a polypeptide chain. Acetylation sites play many important roles, including regulating membrane protein functions and strongly affecting the membrane interaction of proteins and membrane remodeling. Because of these properties, its correct identification is essential to understand its mechanism in biological systems. As such, some traditional methods, such as mass spectrometry and site-directed mutagenesis, are used, but they are tedious and time-consuming. To overcome such limitations, many computer models are being developed to correctly identify their sequences from non-acetyl sequences, but they have poor efficiency in terms of accuracy, sensitivity, and specificity. This work proposes an efficient and accurate computational model for predicting Acetylation using machine learning approaches. The proposed model achieved an accuracy of 100 percent with the 10-fold cross-validation test based on the Random Forest classifier, along with a feature extraction approach using statistical moments. The model is also validated by the jackknife, self-consistency, and independent test, which achieved an accuracy of 100, 100, and 97, respectively, results far better as compared to the already existing models available in the literature.

Sensitivity studies toward the retrieval of ice crystal habit distributions inside cirrus clouds from upwelling far infrared spectral radiance observations

In this paper a retrieval procedure of cirrus ice properties, based on the optimal estimation and the Levenberg–Marquardt iterative formula, is extended with a suitable non linear transformation of the cloud state vector parameters to include the crystal habit distribution in addition to the effective diameter and the optical depth. The cloud optical coefficients are linearly combined using different crystal shapes (solid columns, droxtals, hollow columns, bullet rosettes, aggregates and plates) to build up two different mixtures of habits to be used in the retrieval. This approach allows us to retrieve the coefficients of the habit crystals inside cirrus clouds by using upwelling far infrared spectral radiance measurements. The retrieval capability is shown by performing self-consistency tests using the Simultaneous Atmospheric and Cloud Retrieval (SACR) code and a set of 240 synthetic diverse observations in cloudy-sky generated by using realistic atmospheric scenarios (for mid-latitudes, tropics and poles) and the noise anticipated for the new far infrared sensor Far-infrared Outgoing Radiation Understanding and Monitoring (FORUM), which is planned to be launched as the ninth Earth Explorer (EE-9) by the European Space Agency (ESA) in 2026. The sensitivity of the retrieved crystal habit coefficients to parameters that are not retrieved, such as the surface temperature and emissivity, the cloud top and bottom heights, the ice fraction, the water vapour and temperature profiles, together with the biases caused by the cloud vertical and horizontal inhomogeneity, were also assessed and discussed in the paper.

Evaluation of deep learning techniques for identification of sarcoma-causing carcinogenic mutations.

The abnormal growth of human healthy cells is called cancer. One of the major types of cancer is sarcoma, mostly found in human bones and soft tissue cells. It commonly occurs in children. According to a survey of the United States of America, there are more than 17,000 sarcoma patients registered each year which is 15% of all cancer cases. Recognition of cancer at its early stage saves many lives. The proposed study developed a framework for the early detection of human sarcoma cancer using deep learning Recurrent Neural Network (RNN) algorithms. The DNA of a human cell is made up of 25,000 to 30,000 genes. Each gene is represented by sequences of nucleotides. The nucleotides in a sequence of a driver gene can change which is termed as mutations. Some mutations can cause cancer. There are seven types of a gene whose mutation causes sarcoma cancer. The study uses the dataset which has been taken from more than 134 samples and includes 141 mutations in 8 driver genes. On these gene sequences RNN algorithms Long and Short-Term Memory (LSTM), Gated Recurrent Units and Bi-directional LSTM (Bi-LSTM) are used for training. Rigorous testing techniques such as Self-consistency testing, independent set testing, 10-fold cross-validation test are applied for the validation of results. These validation techniques yield several metrics such as Area Under the Curve (AUC), sensitivity, specificity, Mathew's correlation coefficient, loss, and accuracy. The proposed algorithm exhibits an accuracy of 99.6% with an AUC value of 1.00.

Geometry versus growth

We carry out a multi-probe self-consistency test of the flat Lambda Cold Dark Matter (ΛCDM) model with the aim of exploring potential causes of the reported tensions between high- and low-redshift cosmological observations. We divide the model into two theory regimes determined by the smooth background (geometry) and the evolution of matter density fluctuations (growth), each governed by an independent set of ΛCDM cosmological parameters. This extended model is constrained by a combination of weak gravitational lensing measurements from the Kilo-Degree Survey, galaxy clustering signatures extracted from Sloan Digital Sky Survey campaigns and the Six-Degree Field Galaxy Survey, and the angular baryon acoustic scale and the primordial scalar fluctuation power spectrum measured in Planck cosmic microwave background (CMB) data. For both the weak lensing data set individually and the combined probes, we find strong consistency between the geometry and growth parameters, as well as with the posterior of standard ΛCDM analysis. In the non-split analysis, for which one single set of parameters was used, tension in the amplitude of matter density fluctuations as measured by the parameter S8 persists at around 3σ, with a 1.5% constraint of S8 = 0.776−0.008+0.016 for the combined probes. We also observe a less significant preference (at least 2σ) for higher values of the Hubble constant, H0 = 70.5−1.5+0.7 km s−1 Mpc−1, as well as for lower values of the total matter density parameter Ωm = 0.289−0.005+0.007 compared to the full Planck analysis. Including the subset of the CMB information in the probe combination enhances these differences rather than alleviate them, which we link to the discrepancy between low and high multipoles in Planck data. Our geometry versus growth analysis does not yet yield clear signs regarding whether the origin of the discrepancies lies in ΛCDM structure growth or expansion history but holds promise as an insightful test for forthcoming, more powerful data.