ML Methods: Biomedical Applications
Here's the compiled list in a readable HTML format:
| Application | Field | Machine Learning Method | Inputs | Outputs |
|---|---|---|---|---|
| Virtual screening | Drug design | Convolutional Neural Networks | Chemical structures, receptors | Potential drug candidates |
| De novo drug design | Drug design | Generative Adversarial Networks | Chemical space exploration | Novel drug candidates |
| Drug-target interaction prediction | Drug design | Graph Neural Networks | Molecular structures, proteins | Binding affinity prediction |
| Protein structure prediction | Drug design | Deep Learning | Amino acid sequences | Protein structure prediction |
| QSAR/QSPR modeling | Drug design | Random Forest, Support Vector Machines | Molecular descriptors | Quantitative activity/property prediction |
| Drug repurposing | Drug design | Network-based approaches | Drug-disease associations | Potential new indications |
| Pharmacophore modeling | Drug design | Machine Learning | Ligand-receptor interactions | Pharmacophore models |
| Toxicity prediction | Drug design | Deep Learning | Chemical structures, descriptors | Toxicity prediction |
| Drug formulation optimization | Drug design | Genetic Algorithms, Neural Networks | Formulation parameters | Optimized drug formulation |
| Genomic medicine | Healthcare | Support Vector Machines | Genomic data | Disease risk prediction |
| Predicting disease risk | Healthcare | Logistic Regression, Random Forest | Clinical data | Disease risk prediction |
| Personalized medicine | Healthcare | Decision Trees, Neural Networks | Genomic data, clinical data | Personalized treatment recommendations |
| Clinical decision support systems | Healthcare | Rule-based systems, Neural Networks | Patient data, clinical guidelines | Treatment recommendations |
| Disease diagnosis | Healthcare | Convolutional Neural Networks | Medical images | Disease diagnosis |
| Electronic health records analysis | Healthcare | Natural Language Processing | Electronic health records | Extracted medical information |
| Medical image analysis | Healthcare | Convolutional Neural Networks | Medical images | Image-based diagnosis |
| Health monitoring and wearables | Healthcare | Sensor data analysis | Sensor data | Health status monitoring |
| Drug dosage optimization | Healthcare | Reinforcement Learning | Patient data, treatment history | Optimal drug dosage |
| Predictive modeling for patient outcomes | Healthcare | Recurrent Neural Networks | Patient data, clinical variables | Patient outcome prediction |
| Biosignal processing | Biomedical engineering | Signal Processing, Neural Networks | Physiological signals | Extracted features |
| Medical device design optimization | Biomedical engineering | Genetic Algorithms, Finite Element Analysis | Device parameters | Optimized device design |
| Tissue engineering | Biomedical engineering | Neural Networks, Genetic Algorithms | Cellular and matrix parameters | Tissue growth and morphology |
| Rehabilitation robotics | Biomedical engineering | Reinforcement Learning, Gaussian Processes | Sensor data, user input | Assistive robotic movements |
| Neural prosthetics | Biomedical engineering | Brain-Machine Interfaces, Deep Learning | Neural activity, control signals | Prosthetic limb control</ td> |
| Biomechanics modeling | Biomedical engineering | Finite Element Analysis, Data-Driven Models | Biomechanical data | Predicted tissue/organ behavior |
| Health informatics | Biomedical engineering | Data Mining, Machine Learning | Health records, sensor data | Patterns and insights |
| Medical data mining | Biomedical engineering | Association Rule Mining, Clustering | Health records, genomic data | Knowledge discovery |
| Genetic analysis and genomics | Bioinformatics | Genome Sequencing, Machine Learning | Genomic data | Genetic insights |
| Next-generation sequencing analysis | Bioinformatics | Bioinformatics algorithms | Sequencing data | Genomic variations and annotations |
| Transcriptomics | Bioinformatics | Microarray Analysis, RNA-seq | Gene expression data | Gene expression profiles |
| Proteomics | Bioinformatics | Mass Spectrometry, Machine Learning | Protein data | Protein identification |
| Metagenomics | Bioinformatics | Metagenomic analysis | Microbiome data | Taxonomic and functional profiling |
| Pathway analysis | Bioinformatics | Enrichment Analysis, Network Analysis | Genomic/proteomic data | Pathway enrichment results |
| Variant calling | Bioinformatics | Statistical models, Machine Learning | Sequencing data | Identified genetic variations |
| Comparative genomics | Bioinformatics | Multiple Sequence Alignment | Genomic data | Evolutionary relationships |
| Phylogenetic analysis | Bioinformatics | Phylogenetic algorithms | Sequence data | Evolutionary tree structures |
| Structural bioinformatics | Bioinformatics | Homology Modeling, Docking | Protein structures | Molecular interactions |
| Epigenomics | Bioinformatics | DNA Methylation Analysis | Epigenomic data | Epigenetic patterns |
| Single-cell analysis | Bioinformatics | Single-cell RNA-seq analysis | Single-cell data | Cell heterogeneity analysis |
| Clinical trial design | Drug development | Bayesian Optimization, Decision Trees | Clinical trial parameters | Optimized trial design |
| Patient recruitment | Drug development | Natural Language Processing | Electronic health records, trial criteria | Eligible patient identification |
| Adverse event detection | Drug development | Text Mining, Machine Learning | Adverse event reports | Adverse event identification |
| Drug safety monitoring | Drug development | Signal Detection Methods | Pharmacovigilance data | Safety signal detection |
| Real-world evidence analysis | Drug development | Observational Study Analysis | Real-world patient data | Treatment effectiveness assessment |
| Drug manufacturing optimization | Drug development | Process Optimization, Statistical Modeling | Manufacturing data | Optimized production processes |
| Drug supply chain management | Drug development | Machine Learning, Blockchain | Supply chain data | Transparent and efficient management |
| Clinical trial outcome prediction | Drug development | Random Forest, Support Vector Machines | Clinical trial data | Outcome prediction |
| Pharmacovigilance | Drug development | Text Mining, Association Rules | Adverse event data | Safety signal detection |
| Precision medicine clinical trials | Drug development | Genomic Analysis, Statistical Modeling | Genomic data, patient profiles | Patient stratification |
| Drug response prediction | Drug development | Random Forest, Gradient Boosting | Patient data, drug features | Drug response prediction |
| Cancer detection and diagnosis | Oncology | Deep Learning, Image Analysis | Medical images | Cancer detection and diagnosis |
| Tumor classification | Oncology | Machine Learning, Statistical Modeling | Tumor data | Tumor type classification |
| Treatment response prediction | Oncology | Support Vector Machines | Patient data, treatment history | Treatment response prediction |
| Radiomics | Oncology | Feature Extraction, Machine Learning | Medical images | Radiomic features |
| Radiotherapy treatment planning | Oncology | Optimization Algorithms, Image Analysis | Medical images | Treatment plan optimization |
| Radiogenomics | Oncology | Integration of Imaging and Genomic Data | Imaging data, genomic data | Imaging-genomic associations |
| Biomarker discovery | Oncology | Genomic Analysis, Machine Learning | Genomic data | Predictive biomarker identification |
| Cancer genomics | Oncology | High-throughput Sequencing, Data Mining | Genomic data | Genomic alterations in cancer |
| Drug resistance prediction | Oncology | Deep Learning, Feature Selection | Patient data, drug features | Drug resistance prediction |
| Immunotherapy response prediction | Oncology | Immune Profiling, Machine Learning | Immune-related data | Immunotherapy response prediction |
| Clinical data integration | Data integration | Data Fusion, Ontology-based Integration | Heterogeneous data | Integrated and harmonized data |
| Data preprocessing and cleaning | Data integration | Data Cleaning Techniques | Raw data | Cleaned and preprocessed data |
| Feature selection and extraction | Data integration | Feature Selection Algorithms | High-dimensional data | Relevant feature subsets |
| Predictive modeling and analytics | Data integration | Regression, Classification | Training data, prediction variables | Prediction models |
| Data visualization and interpretation | Data integration | Data Visualization Libraries</td > | Analyzed data | Visual representations |
| Natural language processing | Data integration | Text Mining, Language Models | Textual data | Extracted information |
| Anomaly detection | Data integration | Statistical Modeling, Machine Learning | Data patterns | Detected anomalies |
| Data privacy and security | Data integration | Cryptography, Anonymization | Sensitive data | Privacy-preserving techniques |
| Patient outcome prediction | Predictive analytics | Neural Networks, Decision Trees | Patient data, clinical variables | Predicted patient outcomes |
| Disease progression modeling | Predictive analytics | Markov Models, Hidden Markov Models | Temporal patient data | Disease progression modeling |
| Hospital readmission prediction | Predictive analytics | Logistic Regression, Random Forest | Patient data, clinical variables | Readmission prediction |
| Disease outbreak detection | Predictive analytics | Time Series Analysis, Machine Learning | Epidemiological data | Outbreak detection |
| Predictive maintenance | Predictive analytics | Anomaly Detection, Machine Learning | Equipment sensor data | Equipment failure prediction |
| Fraud detection | Predictive analytics | Anomaly Detection, Machine Learning | Transaction data | Fraudulent activity detection |
| Risk stratification | Predictive analytics | Clustering, Survival Analysis | Patient data, clinical variables | Risk stratification |
| Disease surveillance | Predictive analytics | Time Series Analysis, Machine Learning | Epidemiological data | Disease incidence and trends |
| Adherence prediction | Predictive analytics | Long Short-Term Memory Networks | Patient data, treatment history | Medication adherence prediction |
| Patient monitoring and early warning | Predictive analytics | Time Series Analysis, Machine Learning | Patient vital signs | Early warning of deteriorating health |
| Mental health diagnosis | Mental health | Deep Learning, Natural Language Processing | Patient data, clinical variables | Mental health diagnosis |
| Sentiment analysis | Mental health | Natural Language Processing | Textual data | Sentiment classification |
| Therapy recommendation | Mental health | Collaborative Filtering, Reinforcement Learning | Patient data | Treatment recommendations |
| Suicide risk assessment | Mental health | Machine Learning, Clinical Assessments | Patient data | Suicide risk assessment |
| Emotion recognition | Mental health | Deep Learning, Facial Expression Analysis | Facial expressions | Emotion classification |
| Sleep disorder diagnosis | Mental health | Signal Processing, Machine Learning | Sleep data | Sleep disorder diagnosis |
| Autism spectrum disorder diagnosis | Mental health | Machine Learning, Behavioral Assessments | Patient data | Autism diagnosis |
| Substance abuse prediction | Mental health | Predictive Modeling, Text Mining | Patient data, text data | Substance abuse prediction |
| Cognitive impairment detection | Mental health | Cognitive Assessments, Machine Learning | Cognitive test data | Cognitive impairment detection |
| Virtual reality therapy | Mental health | Virtual Reality, Biofeedback | Patient responses, physiological signals | Therapeutic interventions |
| Rehabilitation assessment | Rehabilitation | Machine Learning, Clinical Assessments | Patient data, assessments | Rehabilitation assessment |
| Assistive technology development | Rehabilitation | Robotics, Machine Learning | Sensor data, user input | Assistive technology control |
| Gait analysis | Rehabilitation | Computer Vision, Machine Learning | Gait data | Gait analysis results |
| Prosthetic limb control | Rehabilitation | Neural Networks, Signal Processing | Neural activity, user input | Prosthetic limb control |
| Stroke rehabilitation | Rehabilitation | Robotics, Virtual Reality | Patient movements, sensor data | Rehabilitation progress |
| Brain-computer interfaces | Rehabilitation | Electroencephalography, Machine Learning | Neural activity, control signals | Brain-computer interaction |
| Spinal cord injury rehabilitation | Rehabilitation | Electrical Stimulation, Machine Learning | Patient sensor data | Rehabilitation progress |
| Motor function assessment | Rehabilitation | Machine Learning, Clinical Assessments | Patient data, assessments | Motor function assessment |
| Mobility prediction | Rehabilitation | Machine Learning, Sensor Fusion | Sensor data | Predicted mobility outcomes |
| Adaptive robotics | Rehabilitation | Reinforcement Learning, Neural Networks | Sensor data, user input | Adaptive robotic movements |
LIST OF ML METHODS:
- Convolutional Neural Networks (CNN)
- Support Vector Machines (SVM)
- Random Forest
- Deep Learning
- Autoencoders
- Generative Adversarial Networks (GAN)
- Graph Neural Networks (GNN)
- Reinforcement Learning (RL)
- Variational Autoencoders (VAE)
- Decision Trees
- Bayesian Networks
- Genetic Algorithms
- Physiologically-Based Pharmacokinetic Models (PBPK)
- Clustering Algorithms
- Recurrent Neural Networks (RNN)
- Particle Swarm Optimization (PSO)
- Artificial Neural Networks (ANN)
- Bayesian Optimization
- Gaussian Processes
- Multiple Linear Regression
- Natural Language Processing (NLP)
- Text Mining
- Sentiment Analysis
- Process Simulation
- Synergy Analysis
- Transfer Learning
- Logistic Regression
- Time Series Analysis
- Anomaly Detection
- Named Entity Recognition (NER)
- Word Embeddings
- Hidden Markov Models (HMM)
- Sensor Fusion
Table:
FOCUSED ON DRUG DESIGN:
Virtual screening:
- Machine Learning Models: Convolutional Neural Networks, Support Vector Machines, Random Forest, Deep Learning, Autoencoders, Generative Adversarial Networks, Graph Neural Networks
- Inputs: Chemical structures, receptor data, ligand-receptor interactions
- Outputs: Potential drug candidates, binding affinity predictions, molecular docking scores
De novo drug design:
- Machine Learning Models: Reinforcement Learning, Generative Adversarial Networks, Variational Autoencoders, Graph Neural Networks
- Inputs: Chemical space exploration, molecular property constraints
- Outputs: Novel drug candidates, molecular structures
Drug-target interaction prediction:
- Machine Learning Models: Graph Neural Networks, Deep Learning, Support Vector Machines, Random Forest
- Inputs: Molecular structures, protein sequences, protein structures, chemical features
- Outputs: Predicted binding affinity, interaction probabilities, drug-target pairs
Protein structure prediction:
- Machine Learning Models: Deep Learning, Convolutional Neural Networks, Recurrent Neural Networks, Generative Models
- Inputs: Amino acid sequences, evolutionary profiles, structural templates
- Outputs: Protein structure models, secondary structure predictions, domain boundaries
QSAR/QSPR modeling:
- Machine Learning Models: Support Vector Machines, Random Forest, Neural Networks, Gradient Boosting, Gaussian Processes
- Inputs: Molecular descriptors, physicochemical properties, molecular fingerprints
- Outputs: Activity/property predictions, quantitative structure-activity relationships
Drug repurposing:
- Machine Learning Models: Network-based approaches, Matrix Factorization, Graph Mining, Deep Learning
- Inputs: Drug-disease associations, biological networks, molecular structures, clinical data
- Outputs: Potential new indications, drug-disease interaction predictions
Pharmacophore modeling:
- Machine Learning Models: Machine Learning, Genetic Algorithms, Evolutionary Algorithms
- Inputs: Ligand-receptor interactions, chemical features, molecular structures
- Outputs: Pharmacophore models, ligand binding preferences, molecular recognition patterns
Toxicity prediction:
- Machine Learning Models: Deep Learning, Random Forest, Support Vector Machines, Bayesian Networks
- Inputs: Chemical structures, molecular descriptors, toxicological data
- Outputs: Toxicity predictions, risk assessments, safety profiles
Drug formulation optimization:
- Machine Learning Models: Genetic Algorithms, Particle Swarm Optimization, Artificial Neural Networks, Bayesian Optimization
- Inputs: Formulation parameters, physicochemical properties, drug delivery profiles
- Outputs: Optimized drug formulation, drug release kinetics, dosage forms
Drug-target interaction network analysis:
- Machine Learning Models: Network Analysis, Graph Neural Networks, Deep Learning, Clustering Algorithms
- Inputs: Drug-target interaction data, biological networks, molecular structures
- Outputs: Identification of drug-target communities, network-based drug repurposing, drug-target interaction predictions
Pharmacokinetics modeling:
- Machine Learning Models: Physiologically-Based Pharmacokinetic Models, Bayesian Networks, Deep Learning
- Inputs: Physicochemical properties, drug metabolism data, physiological parameters
- Outputs: Drug absorption, distribution, metabolism, and excretion predictions, plasma concentration profiles
Molecular docking:
- Machine Learning Models: Genetic Algorithms, Convolutional Neural Networks, Reinforcement Learning
- Inputs: Protein structures, ligand structures, binding site information
- Outputs: Docked complex conformations, binding affinity predictions, protein-ligand interaction patterns
Drug solubility prediction:
- Machine Learning Models: Random Forest, Support Vector Machines, Artificial Neural Networks, Deep Learning
- Inputs: Molecular descriptors, chemical structures, physicochemical properties
- Outputs: Solubility predictions, solubility class assignments, formulation recommendations
ADME (Absorption, Distribution, Metabolism, Excretion) property prediction:
- Machine Learning Models: Support Vector Machines, Random Forest, Neural Networks, Gradient Boosting
- Inputs: Molecular descriptors, physicochemical properties, structural features
- Outputs: Absorption, distribution, metabolism, and excretion property predictions, compound selection for further development
Quantitative structure-activity relationship (QSAR) modeling:
- Machine Learning Models: Multiple Linear Regression, Support Vector Machines, Random Forest, Neural Networks
- Inputs: Molecular descriptors, biological activity data, chemical structures
- Outputs: Structure-activity relationship models, activity predictions, compound prioritization
Drug delivery system design:
- Machine Learning Models: Optimization Algorithms, Reinforcement Learning, Genetic Algorithms
- Inputs: Drug properties, formulation parameters, release kinetics, target tissues
- Outputs: Optimal drug delivery systems, release profiles, targeting strategies
Pharmacovigilance:
- Machine Learning Models: Natural Language Processing, Text Mining, Sentiment Analysis, Deep Learning
- Inputs: Adverse event reports, electronic health records, social media data
- Outputs: Detection of adverse drug reactions, signal identification, safety surveillance
Pharmaceutical process optimization:
- Machine Learning Models: Genetic Algorithms, Process Simulation, Neural Networks, Reinforcement Learning
- Inputs: Process parameters, quality control data, manufacturing variables
- Outputs: Optimized process parameters, yield improvement, cost reduction
Drug combination prediction:
- Machine Learning Models: Synergy Analysis, Deep Learning, Random Forest, Matrix Factorization
- Inputs: Drug-target interactions, chemical structures, gene expression data
- Outputs: Predicted synergistic drug combinations, combination response predictions, personalized combination therapies
OLD
Drug Design:
- Virtual screening: Convolutional Neural Networks (CNN), Chemical structures, receptors, Potential drug candidates.
- De novo drug design: Generative Adversarial Networks (GAN), Chemical space exploration, Novel drug candidates.
- Drug-target interaction prediction: Graph Neural Networks (GNN), Molecular structures, proteins, Binding affinity prediction.
- Protein structure prediction: Deep Learning, Amino acid sequences, Protein structure prediction.
- QSAR/QSPR modeling: Random Forest, Support Vector Machines, Molecular descriptors, Quantitative activity/property prediction.
- Drug repurposing: Network-based approaches, Drug-disease associations, Potential new indications.
- Pharmacophore modeling: Machine Learning, Ligand-receptor interactions, Pharmacophore models.
- Toxicity prediction: Deep Learning, Chemical structures, descriptors, Toxicity prediction.
- Drug formulation optimization: Genetic Algorithms, Neural Networks, Formulation parameters, Optimized drug formulation.
Healthcare:
- Genomic medicine: Support Vector Machines, Genomic data, Disease risk prediction.
- Predicting disease risk: Logistic Regression, Random Forest, Clinical data, Disease risk prediction.
- Personalized medicine: Decision Trees, Neural Networks, Genomic data, clinical data, Personalized treatment recommendations.
- Clinical decision support systems: Rule-based systems, Neural Networks, Patient data, clinical guidelines, Treatment recommendations.
- Disease diagnosis: Convolutional Neural Networks (CNN), Medical images, Disease diagnosis.
- Electronic health records analysis: Natural Language Processing (NLP), Electronic health records, Extracted medical information.
- Medical image analysis: Convolutional Neural Networks (CNN), Medical images, Image-based diagnosis.
- Health monitoring and wearables: Sensor data analysis, Sensor data, Health status monitoring.
- Drug dosage optimization: Reinforcement Learning, Patient data, treatment history, Optimal drug dosage.
- Predictive modeling for patient outcomes: Recurrent Neural Networks (RNN), Patient data, clinical variables, Patient outcome prediction.
Biomedical Engineering:
- Biosignal processing: Signal Processing, Neural Networks, Physiological signals, Extracted features.
- Medical device design optimization: Genetic Algorithms, Finite Element Analysis, Device parameters, Optimized device design.
- Tissue engineering: Neural Networks, Genetic Algorithms, Cellular and matrix parameters, Tissue growth and morphology.
- Rehabilitation robotics: Reinforcement Learning, Gaussian Processes, Sensor data, user input, Assistive robotic movements.
- Neural prosthetics: Brain-Machine Interfaces, Deep Learning, Neural activity, control signals, Prosthetic limb control.
- Biomechanics modeling: Finite Element Analysis, Data-Driven Models, Biomechanical data, Predicted tissue/organ behavior.
- Health informatics: Data Mining, Machine Learning, Health records, sensor data, Patterns and insights.
- Medical data mining: Association Rule Mining, Clustering, Health records, genomic data, Knowledge discovery.
Bioinformatics:
- Genetic analysis and genomics: Genome Sequencing, Machine Learning, Genomic data, Genetic insights.
- Next-generation sequencing analysis: Bioinformatics algorithms, Sequencing data, Genomic variations and annotations.
- Transcriptomics: Microarray Analysis, RNA-seq, Gene expression data, Gene expression profiles.
- Proteomics: Mass Spectrometry, Machine Learning, Protein data, Protein identification.
- Metagenomics: Metagenomic analysis, Microbiome data, Taxonomic and functional profiling.
- Pathway analysis: Enrichment Analysis, Network Analysis, Genomic/proteomic data, Pathway enrichment results.
- Variant calling: Statistical models, Machine Learning, Sequencing data, Identified genetic variations.
- Comparative genomics: Multiple Sequence Alignment, Genomic data, Evolutionary relationships.
- Phylogenetic analysis: Phylogenetic algorithms, Sequence data, Evolutionary tree structures.
- Structural bioinformatics: Homology Modeling, Docking, Protein structures, Molecular interactions.
- Epigenomics: DNA Methylation Analysis, Epigenomic data, Epigenetic patterns.
- Single-cell analysis: Single-cell RNA-seq analysis, Single-cell data, Cell heterogeneity analysis.
Drug Development:
- Clinical trial design: Bayesian Optimization, Decision Trees, Clinical trial parameters, Optimized trial design.
- Patient recruitment: Natural Language Processing (NLP), Electronic health records, trial criteria, Eligible patient identification.
- Adverse event detection: Text Mining, Machine Learning, Adverse event reports, Adverse event identification.
- Drug safety monitoring: Signal Detection Methods, Pharmacovigilance data, Safety signal detection.
- Real-world evidence analysis: Observational Study Analysis, Real-world patient data, Treatment effectiveness assessment.
- Drug manufacturing optimization: Process Optimization, Statistical Modeling, Manufacturing data, Optimized production processes.
- Drug supply chain management: Machine Learning, Blockchain, Supply chain data, Transparent and efficient management.
- Clinical trial outcome prediction: Random Forest, Support Vector Machines, Clinical trial data, Outcome prediction.
- Pharmacovigilance: Text Mining, Association Rules, Adverse event data, Safety signal detection.
- Precision medicine clinical trials: Genomic Analysis, Statistical Modeling, Genomic data, patient profiles, Patient stratification.
- Drug response prediction: Random Forest, Gradient Boosting, Patient data, drug features, Drug response prediction.
Oncology:
- Cancer detection and diagnosis: Deep Learning, Image Analysis, Medical images, Cancer detection and diagnosis.
- Tumor classification: Machine Learning, Statistical Modeling, Tumor data, Tumor type classification.
- Treatment response prediction: Support Vector Machines, Patient data, treatment history, Treatment response prediction.
- Radiomics: Feature Extraction, Machine Learning, Medical images, Radiomic features.
- Radiotherapy treatment planning: Optimization Algorithms, Image Analysis, Medical images, Treatment plan optimization.
- Radiogenomics: Integration of Imaging and Genomic Data, Imaging data, genomic data, Imaging-genomic associations.
- Biomarker discovery: Genomic Analysis, Machine Learning, Genomic data, Predictive biomarker identification.
- Cancer genomics: High-throughput Sequencing, Data Mining, Genomic data, Genomic alterations in cancer.
- Drug resistance prediction: Deep Learning, Feature Selection, Patient data, drug features, Drug resistance prediction.
- Immunotherapy response prediction: Immune Profiling, Machine Learning, Immune-related data, Immunotherapy response prediction.