Multiple Layer Perceptron Research Articles

Estimating heavy metals absorption efficiency in an aqueous solution using nanotube-type halloysite from weathered pegmatites and a novel Harris hawks optimization-based multiple layers perceptron neural network

Estimating heavy metals absorption efficiency in an aqueous solution using nanotube-type halloysite from weathered pegmatites and a novel Harris hawks optimization-based multiple layers perceptron neural network

Application of artificial neural network (ANN-MLP) for the prediction of fouling resistance in heat exchanger to MgO-water and CuO-water nanofluids.

In this work, an artificial neural network (ANN) model was developed with the aim of predicting fouling resistance for heat exchanger, the network was designed and trained by means of 375 experimental data points that were selected from the literature. These data points contain six inputs, including time, volumetric concentration, heat flux, mass flow rate, inlet temperature, thermal conductivity and fouling resistance as an output. The experimental data are used for training, testing and validation of the ANN using multiple layer perceptron (MLP). The comparison of statistical criteria of different networks shows that the optimal structure for predicting the fouling resistance of the nanofluid is the MLP network with 20 hidden neurons, which has been trained with Levenberg-Marquardt (LM) algorithm. The accuracy of the model was assessed based on three known statistical metrics including mean square error (MSE), mean absolute percentage error (MAPE) and coefficient of determination (R2). The obtained model was found with the performance of {MSE = 6.5377 × 10-4, MAPE = 2.40% and R2 = 0.99756} for the training stage, {MSE = 3.9629 × 10-4, MAPE = 1.8922% and R2 = 0.99835} for the test stage and {MSE = 5.8303 × 10-4, MAPE = 2.57% and R2 = 0.99812} for the validation stage. In order to control the fouling procedure, and after conducting a sensitivity analysis, it found that all input variables have strong effect on the estimation of the fouling resistance.

A deep learning-based computer-aided diagnosis method of X-ray images for bone age assessment

Bone age assessment using hand-wrist X-ray images is fundamental when diagnosing growth disorders of a child or providing a more patient-specific treatment. However, as clinical procedures are a subjective assessment, the accuracy depends highly on the doctor’s experience. Motivated by this, a deep learning-based computer-aided diagnosis method was proposed for performing bone age assessment. Inspired by clinical approaches and aimed to reduce expensive manual annotations, informative regions localization based on a complete unsupervised learning method was firstly performed and an image-processing pipeline was proposed. Subsequently, an image model with pre-trained weights as a backbone was utilized to enhance the reliability of prediction. The prediction head was implemented by a Multiple Layer Perceptron with one hidden layer. In compliance with clinical studies, gender information was an additional input to the prediction head by embedded into the feature vector calculated from the backbone model. After the experimental comparison study, the best results showed a mean absolute error of 6.2 months on the public RSNA dataset and 5.1 months on the additional dataset using MobileNetV3 as the backbone.

A Tandem Feature Extraction Approach for Arrhythmia Identification

Heart disease is currently the leading cause of death in the world. The electrocardiogram (ECG) is the recording of the electrical activity generated by the heart. Its low cost and simplicity have made it an essential test for monitoring heart disease, especially for the identification of arrhythmias. With the advances in electronic technology, there are nowadays sensors that enable the recording of the ECG during the daily life of the patient and its wireless transmission to healthcare facilities. This type of information has a great potential to detect cardiac diseases in their early stages and to permit early interventions before the patient’s health deteriorates. However, to usefully exploit the large volume of information obtained from ambulatory ECG, pattern recognition techniques that are capable of automatically analyzing it are required. Tandem feature extraction techniques have proven to be useful for the processing of physiological parameters such as the electroencephalogram (EEG) and speech. However, to the best of our knowledge, they have never been applied to the ECG. In this paper, the utility of tandem feature extraction for the identification of arrhythmias is studied. The coefficients of a regression using Hermite functions are used to create a feature vector that represents the heartbeat. A multiple-layer perceptron (MLP) is trained using these features and its posterior probability outputs are used to extend the original feature vector. Finally, a Gaussian mixture model (GMM) is trained on the extended feature vectors, which is then used in a GMM-based arrhythmia identification system. This approach has been validated using the MIT-BIH Arrhythmia database. The accuracy of the Gaussian mixture model increased by 15.8% when applied over the extended feature vectors, compared to its application over the original feature vectors, showing the potential of tandem feature extraction for ECG analysis and arrhythmia identification.

GanDTI: A multi-task neural network for drug-target interaction prediction

Drug discovery processes require drug-target interaction (DTI) prediction by virtual screenings with high accuracy. Compared with traditional methods, the deep learning method requires less time and domain expertise, while achieving higher accuracy. However, there is still room for improvement for higher performance with simplified structures. Meanwhile, this field is calling for multi-task models to solve different tasks. Here we report the GanDTI, an end-to-end deep learning model for both interaction classification and binding affinity prediction tasks. This model employs the compound graph and protein sequence data. It only consists of a graph neural network, an attention module and a multiple-layer perceptron, yet outperforms the state-of-the art methods to predict binding affinity and interaction classification on the DUD-E, human, and bindingDB benchmark datasets. This demonstrates our refined model is highly effective and efficient for DTI prediction and provides a new strategy for performance improvement.

Prediction of heavy metals removal by polymer inclusion membranes using machine learning techniques

AbstractThis study predicts heavy metals removal from aqueous solution by polymer inclusion membranes (PIMs) process using machine learning (ML) techniques such as multiple layer perceptron neural networks (MLPNN) and multiple linear regression (MLR) after data analysis. The removal efficiency (RE) of the PIMs process is predicted for cobalt (Co), cadmium (Cd) and chromium (Cr) by changing operating conditions including time, carrier type, carrier rate, film thickness, plasticizer type and plasticizer rate. The MLPNN model presents reliable results with lower mean square error (MSE) for an unseen testing dataset, whereas the MLR model shows higher MSE values. The coefficient of determination (R2) of the MLPNN model for the testing dataset is 0.93, 0.90 and 0.86 for Co, Cd and Cr, respectively, whereas MLR shows poor results. Therefore, the MLPNN model can be a competitive, robust and fast alternate to optimize the PIMs process with minimum experimental work.

Overcoming Long-Term Catastrophic Forgetting Through Adversarial Neural Pruning and Synaptic Consolidation.

Enabling a neural network to sequentially learn multiple tasks is of great significance for expanding the applicability of neural networks in real-world applications. However, artificial neural networks face the well-known problem of catastrophic forgetting. What is worse, the degradation of previously learned skills becomes more severe as the task sequence increases, known as the long-term catastrophic forgetting. It is due to two facts: first, as the model learns more tasks, the intersection of the low-error parameter subspace satisfying for these tasks becomes smaller or even does not exist; second, when the model learns a new task, the cumulative error keeps increasing as the model tries to protect the parameter configuration of previous tasks from interference. Inspired by the memory consolidation mechanism in mammalian brains with synaptic plasticity, we propose a confrontation mechanism in which Adversarial Neural Pruning and synaptic Consolidation (ANPyC) is used to overcome the long-term catastrophic forgetting issue. The neural pruning acts as long-term depression to prune task-irrelevant parameters, while the novel synaptic consolidation acts as long-term potentiation to strengthen task-relevant parameters. During the training, this confrontation achieves a balance in that only crucial parameters remain, and non-significant parameters are freed to learn subsequent tasks. ANPyC avoids forgetting important information and makes the model efficient to learn a large number of tasks. Specifically, the neural pruning iteratively relaxes the current task's parameter conditions to expand the common parameter subspace of the task; the synaptic consolidation strategy, which consists of a structure-aware parameter-importance measurement and an element-wise parameter updating strategy, decreases the cumulative error when learning new tasks. Our approach encourages the synapse to be sparse and polarized, which enables long-term learning and memory. ANPyC exhibits effectiveness and generalization on both image classification and generation tasks with multiple layer perceptron, convolutional neural networks, and generative adversarial networks, and variational autoencoder. The full source code is available at https://github.com/GeoX-Lab/ANPyC.

A generalized artificial intelligence model for estimating the friction angle of clays in evaluating slope stability using a deep neural network and Harris Hawks optimization algorithm

In landslide susceptibility mapping or evaluating slope stability, the shear strength parameters of rocks and soils and their effectiveness are undeniable. However, they have not been studied for all-natural materials, as well as different locations. Therefore, this paper proposes a novel generalized artificial intelligence model for estimating the friction angle of clays from different areas/locations for evaluating slope stability or landslide susceptibility mapping, including the datasets from the UK, New Zealand, Indonesia, Venezuela, USA, Japan, and Italy. The robustness and consistency of the model’s prediction were checked by testing with various datasets having different geological and geomorphological setups. Accordingly, 162 observations from different areas/locations were collected from the locations and regions above for this aim. Subsequently, deep learning techniques were applied to develop the multiple layer perceptron (MLP) neural network model (i.e., DMLP model) with the goal of error reduction of the MLP model. Next, Harris Hawks optimization (HHO) algorithm was applied to boost the optimization of the DMLP model for predicting friction angle of clays aiming to get a better accuracy than those of the DMLP model, called HHO–DMLP model. A DMLP neural network without optimization of the HHO algorithm and two other conventional models (i.e., SVM and RF) were also employed to compare with the proposed HHO–DMLP model. The results showed that the proposed HHO–DMLP model predicted the friction angle of clays better than those of the other models. It can reflect the friction angle of clays with acceptable accuracy from different locations and regions (i.e., MSE = 12.042; RMSE = 3.470; R2 = 0.796; MAPE = 0.182; and VAF = 78.806). The DMLP model without optimization of the HHO algorithm provided slightly lower accuracy (i.e., MSE = 15.151; RMSE = 3.892; R2 = 0.738; MAPE = 0.202; and VAF = 73.431). Besides, two other conventional models (i.e., SVM and RF) provided low reliability, especially over-fitting happened with the RF model, and it was not recommended to be used to predict the friction angle of clays (i.e., RMSE = 6.325 and R2 = 0.377 on the training dataset, but RMSE = 1.669 and R2 = 0.961 on the testing dataset).

Probabilistic pushover analysis of reinforced concrete frame structures using dropout neural network

When taking into consideration nonlinear phenomena such as material plasticity, plastic hinge, and P-Delta effect, the pushover analysis can provide more realistic structures’ nonlinear responses. However, this method is not widely used in practice as it is more complex and requires more expertise than elastic approaches. On the other hand, the data-driven method emerges as an increasingly appealing alternative since it requires only input parameters, then directly yields results in conditions that enough training data are provided, as well as an appropriate machine learning model is devised. Thus, this study develops a probabilistic data-driven approach using the Multiple Layer Perceptron network coupled with the Dropout mechanism to perform the pushover analysis of reinforced concrete (RC) frame structures, predicting base shear, lateral displacement, as well as their relationship between the two formers. Moreover, corresponding confidence intervals of predicted values are also available owing to the probabilistic nature of the method, thus helping engineers design conservative solutions.
 Keywords:
 pushover analysis; reinforced concrete; structure; probabilistic analysis; machine learning; dropout mechanism; OpenSees.

Adaptive diagnostic machine learning technique for classification of cell decisions for AKT protein

Artificial intelligence techniques can unravel clinically consistent information in clinical data which in turn assist in decision-making. With the help of state-of-the-art techniques, diagnostic systems are used to identify various diseases by using different machine learning (ML) techniques. In this paper, a combination of ML techniques such as Radial Basis Function (RBF) and Multiple Layer Perceptron (MLP) was used to predict cell decisions (cell survival/death) of AKT protein. AKT signalling networks have various downstream consequences on cellular metabolism either directly through the regulation of nutrient transporters, metabolic enzymes or indirectly through the control of transcription factors that regulate the expression of metabolic pathways which determine cell survival, cell growth, and cell death. Experimental analysis was performed in this work to examine the signalling networks that determine cell survival/death decisions by using an amalgamation of three proteins for ten different combinations in 13 different slices for a period of 0–24 h. P-P plot, Q-Q plot, and histogram tests were used for data visualization to determine which distribution the data fits. In addition, goodness of fit test was also employed using distribution functions such as Weibull, Exponential, and Normal distribution to determine whether the data fits a distribution of a certain population. The results were validated by calculating the MTTF values. The results of the analysis performed show that the Weibull distribution yields remarkable results. Also the results obtained with the Multiple Layer Perceptron, MLP 10-8-1 was found to perform better than other techniques giving an accuracy of 99.33% when the exponential activation function was used. The results of the experimental study indicate that it is possible to create self-consistent cell-signalling compendia based on AKT protein data that have been computationally simulated to provide valuable insights for cell survival/death regulation.

Multiply accumulate operations in memristor crossbar arrays for analog computing

Memristors are now becoming a prominent candidate to serve as the building blocks of non-von Neumann in-memory computing architectures. By mapping analog numerical matrices into memristor crossbar arrays, efficient multiply accumulate operations can be performed in a massively parallel fashion using the physics mechanisms of Ohm’s law and Kirchhoff’s law. In this brief review, we present the recent progress in two niche applications: neural network accelerators and numerical computing units, mainly focusing on the advances in hardware demonstrations. The former one is regarded as soft computing since it can tolerant some degree of the device and array imperfections. The acceleration of multiple layer perceptrons, convolutional neural networks, generative adversarial networks, and long short-term memory neural networks are described. The latter one is hard computing because the solving of numerical problems requires high-precision devices. Several breakthroughs in memristive equation solvers with improved computation accuracies are highlighted. Besides, other nonvolatile devices with the capability of analog computing are also briefly introduced. Finally, we conclude the review with discussions on the challenges and opportunities for future research toward realizing memristive analog computing machines.

DGCB-Net: Dynamic Graph Convolutional Broad Network for 3D Object Recognition in Point Cloud

3D (3-Dimensional) object recognition is a hot research topic that benefits environment perception, disease diagnosis, and the mobile robot industry. Point clouds collected by range sensors are a popular data structure to represent a 3D object model. This paper proposed a 3D object recognition method named Dynamic Graph Convolutional Broad Network (DGCB-Net) to realize feature extraction and 3D object recognition from the point cloud. DGCB-Net adopts edge convolutional layers constructed by weight-shared multiple-layer perceptrons (MLPs) to extract local features from the point cloud graph structure automatically. Features obtained from all edge convolutional layers are concatenated together to form a feature aggregation. Unlike stacking many layers in-depth, our DGCB-Net employs a broad architecture to extend point cloud feature aggregation flatly. The broad architecture is structured utilizing a flat combining architecture with multiple feature layers and enhancement layers. Both feature layers and enhancement layers concatenate together to further enrich the features’ information of the point cloud. All features work on the object recognition results thus that our DGCB-Net show better recognition performance than other 3D object recognition algorithms on ModelNet10/40 and our scanning point cloud dataset.

Artificial neural network models for prediction of net radiation over a tropical region

In this study, net radiation and temperature data were collected from the Nigeria Meteorological Agency, Lagos, covering a period of 1983–2013. The study uses a multiple layer perceptron (MLP) model of artificial neural network (ANN) with three algorithms namely the Gradient Descent, Conjugate Gradient, and Broyden–Fletcher–Goldfarb–Shanno for the prediction of net radiation using the temperature series as input variables in place of commonly used empirical statistical methods. The investigation is conducted in a tropical region where 16 meteorological stations spatially distributed across the four climatic regions of Nigeria viz semi-arid (SAR), sub-humid dry (SHD), sub-humid humid (SHH), and humid (HUM) regions were used as a case study. Analyses showed that minimum temperature has a greater positive impact on net radiation than maximum temperature. This has been attributed among other factors to have contributed to higher magnitudes of net radiation in humid regions than those of arid regions. Meanwhile, the performance assessment of the ANN models using the refined index (dr) metric showed that among the three models proposed, the MLP-BFGS model performed best having maximum values of 0.91 in the SAR, 0.65 in the SHD, 0.80 in the SHH, and 0.88 in the HUM regions. It also has the lowest error analysis with magnitudes of root mean square error values of 7.92 W/m2 in the SAR, 9.95 W/m2 in the SHD, 12.91 W/m2 in the SHH, and 8.73 W/m2 in the HUM regions. Finally, it can be inferred from the results that MLP neural networks with the BFGS algorithm can predict net radiation accurately better than any empirical statistical methods over Nigeria.

Development of a Risk Prediction Model for Central-Line–Associated Bloodstream Infection (CLABSI) in Patients With Continuous Renal Replacement Therapy

Background: The number of patients with end-stage renal disease and acute kidney injury in China is large and increasing year by year. Continuous renal replacement therapy (CRRT) is one of the important treatment methods. However, long-time CRRT would inevitably lead to CLABSI, which would seriously affect the treatment and prognosis of the patient. Although CLABSIs can be prevented and controlled, the rate of CLABSI in China is still higher than in other countries. Therefore, it is urgent to find new intervention methods on the basis of existing methods. Surveillance is the prerequisite of infection prevention and control. We sought to develop a risk prediction model for CLABSI in patients with CRRT according to uncontrollable risk factors, which could be used for early assessment and screening of high-risk infection groups. Such a tool would bring the supervision and infection control to the forefront in addressing these issues. Methods: We selected 3,103 CRRT patients in the West China Hospital of Sichuan University from January 2013 to December 2018 using the hospital infection and infectious disease monitoring module of electronic medical records (EMR) system with the integration and elimination criteria. Data mining and feature selection were performed using Weka software. Separately, prediction models developed by Weka software and SPSS software were compared with each other using the area under the curve (AUC) method to assess the performance of the forecasting models. Result: The incidence of CLABSI in CRRT patients was 8.01 per 1,000 catheter days (238 of 29,711). According to the multifactor regression analysis by SPSS software, the retaining time of dialysis catheter, C-reactive protein levels, total bilirubin, acute pancreatitis, and systemic inflammation reaction syndrome were the risk factors. According to the Youden’s index, the cutoff point of the retaining time of dialysis catheter was 5.5 days; the cutoff point of CRP was 112.5mg/L; and the cutoff point of total bilirubin was 14.15 μmol/L. The prediction models of CLABSI for CRRT patients were developed: The AUC of the prediction model developed by SPSS software was 0.763 (95% CI, 0.717–0.809). The receiver operating characteristic (ROC) curve analysis showed that the AUCs of the prediction models developed separately by Weka software using Bayes, logistic regression analysis, multiple layer Perceptron and J48, and SPSS software through logistic regression analysis were between 0.6 and 0.8. Using the down-sampling technique, the AUC ranged between 0.7 and 0.9, and the sensitivity, precision, and κ value increased. Thus, these models had definite clinical significance. Conclusion: The prediction models of CLABSI for CRRT patients, developed based on the big healthcare data, not only had good judgment ability, but also had good application value for individual evaluations and the target population.Funding: This study was supported by the Health Commission of Sichuan Province.Disclosures: None

Vulnerability of Hardware Neural Networks to Dynamic Operation Point Variations

Editor’s notes: This article studies the impacts of physical variations on neural networks. The proposed studies reveal an important observation that both multiple layer perceptron (MLP) and convolutional neural network (CNN) may fail to operate appropriately even with small variations (e.g., voltage droops as small as 20 mV). Robust neural network architectures, including the binarized neural network (BNN) and the local binary pattern network (LBPNet), are explored to address this variability issue that has become a major bottleneck for practical applications. — Xin Li, Duke University

Regional landslide hazard assessment through integrating susceptibility index and rainfall process

Due to the difference of the spatial and temporal distribution of rainfall and the complex diversity of the disaster-prone environment (topography, geological, fault, and lithology), it is difficult to assess the hazard of landslides at the regional scale quantitatively only considering rainfall condition. Based on detailed landslide inventory and rainfall data in the hilly area in Sichuan province, this study analyzed the effects of both rainfall process and environmental factors on the occurrence of landslides. Through analyzing environmental factors, a landslide susceptibility index (LSI) was calculated using multiple layer perceptron (MLP) model to reflect the regional landslide susceptibility. Further, the characteristics of rainfall process and landslides were examined quantitatively with statistical analysis. Finally, a probability model integrating LSI and rainfall process was constructed using logistical regression analysis to assess the landslide hazard. Validation showed satisfactory results, and the inclusion of LSI effectively improved the accuracy of the landslide hazard assessment: Compared with only considering the rainfall process factors, the accuracy of the landslide prediction model both considering the rainfall process and landslide susceptibility is improved by 3%. These results indicate that an integration of susceptibility index and rainfall process is essential in improving the timeliness and accuracy of regional landslide early warning.

Using artificial neural network for predicting heat transfer coefficient during flow boiling in an inclined channel

The flow and heat transfer characteristics in a nuclear power plant in the event of a serious accident are simulated by boiling water in an inclined rectangular channel. In this study an artificial neural network model was developed with the aim of predicting heat transfer coefficient for flow boiling of water in inclined channel, the network was designed and trained by means of 520 experimental data points that were selected from within the literature. Orientation,mass flux, quality and heat flow which were employed to serve as variables of input of multiple layer perceptron neural network, whereas the analogous heat transfer coefficient was selected to be its output. Via the method of trial-and-error, multiple layer perceptron network with 30 neurons in the hidden layer was attained as optimal arteficial neural network structure. The fact that is was enabled to predict accurately the heat transfer coefficient. For the training set, the mean relative absolute error is about 0.68 % and the correlation coefficient, is about 0.9997. As for the testing and validation set they are, respectively, about 0.60 % and 0.9998 and about 0.79 % and 0.9996.

Estimation of Pressure Drop of Two-Phase Flow in Horizontal Long Pipes Using Artificial Neural Networks

Abstract Gas–liquid two-phase flows through long pipelines are one of the most common cases found in chemical, oil, and gas industries. In contrast to the gas/Newtonian liquid systems, the pressure drop has rarely been investigated for two-phase gas/non-Newtonian liquid systems in pipe flows. In this regard, an artificial neural networks (ANNs) model is presented by employing a large number of experimental data to predict the pressure drop for a wide range of operating conditions, pipe diameters, and fluid characteristics. Utilizing a multiple-layer perceptron neural network (MLPNN) model, the predicted pressure drop is in a good agreement with the experimental results. In most cases, the deviation of the predicted pressure drop from the experimental data does not exceed 5%. It is observed that the MLPNN provides more accurate results for horizontal pipelines in comparison with other empirical correlations that are commonly used in industrial applications.

Landslide susceptibility prediction based on a semi-supervised multiple-layer perceptron model

Conventional supervised and unsupervised machine learning models used for landslide susceptibility prediction (LSP) have many drawbacks, such as an insufficient number of recorded landslide samples, and the subjective and random selection of non-landslide samples. To overcome these drawbacks, a semi-supervised multiple-layer perceptron (SSMLP) is innovatively proposed with several processes: (1) an initial landslide susceptibility map (LSM) is produced using the multiple-layer perceptron (MLP) based on the original recorded landslide samples and related environmental factors; (2) the initial LSM is respectively classified into five areas with very high, high, moderate, low and very low susceptible levels; (3) some reasonable grid units from the areas with very high susceptible level are selected as new landslide samples to expand the original landslide samples; (4) reasonable non-landslide samples are selected from the areas with very low susceptible level; and (5) the expanded landslide samples, reasonable selected non-landslide samples and related environmental factors are put into the MLP once again to predict the final LSM. The Xunwu County of Jiangxi Province in China is selected as the study area. Conventional supervised machine learning (i.e. MLP) and unsupervised machine learning (i.e. K-means clustering model) are selected for comparisons. The comparative results indicate that the SSMLP model has a considerably higher LSP performance than the MLP and K-means clustering in Xunwu County. The SSMLP model successfully addresses the drawbacks existed in the conventional machine learning for LSP.

Computational modeling and experimental analysis for the diagnosis of cell survival/death for Akt protein

BackgroundSignalling systems that control cell decisions allow cells to process input signals by apprehending the information of the cell to give one of these two feasible outputs: cell death or cell survival. In this paper, a well-structured control design methodology supported by a hierarchical design system was developed to examine signalling networks that control cell decisions by considering a combinations of three primary signals (input proteins): the pro survival growth factors, epidermal growth factor (EGF), insulin, and the pro death cytokine, tumour necrosis factor-α (TNF), for AKT/protein kinase B. The AKT actions were examined by using the three input proteins for cell survival/apoptosis for a period of 0–24 h in 13 different slices for ten different combinations. ResultsExperimental analysis was performed to consider the reactions that were essential to explain the action of AKT. Furthermore, pre-processing and data normalization were performed by using standard deviation, plotting histograms, and scatter plots. Feature extraction and selection were performed using correlation matrix. Radial basis function (RBF) and multiple-layer perceptron (MLP) were used for cell survival/death classification. For all the ten combinations of the three input proteins, 42.85, 347.22, 153.13 were obtained as the minimum value, maximum value, and mean value, respectively, and 126.11 was obtained as the standard deviation for 5-0-5 ng/ml combinations of TNF-EGF-Insulin. The results obtained with MLP 10-8-1 were found to outperform other techniques. ConclusionThe results from the experimental analysis indicate that it is possible to build self-consistent compendia cell-signalling data based on AKT protein which were simulated computationally to yield important insights for the control of cell survival/death.