Prediction Pipeline Research Articles

Accelerating machine learning queries with linear algebra query processing

The rapid growth of large-scale machine learning (ML) models has led numerous commercial companies to utilize ML models for generating predictive results to help business decision-making. As two primary components in traditional predictive pipelines, data processing, and model predictions often operate in separate execution environments, leading to redundant engineering and computations. Additionally, the diverging mathematical foundations of data processing and machine learning hinder cross-optimizations by combining these two components, thereby overlooking potential opportunities to expedite predictive pipelines. In this paper, we propose an operator fusion method based on GPU-accelerated linear algebraic evaluation of relational queries. Our method leverages linear algebra computation properties to merge operators in machine learning predictions and data processing, significantly accelerating predictive pipelines by up to 317x. We perform a complexity analysis to deliver quantitative insights into the advantages of operator fusion, considering various data and model dimensions. Furthermore, we extensively evaluate linear algebra query processing and operator fusion utilizing the widely-used Star Schema and TPC-DI benchmarks. Through comprehensive evaluations, we demonstrate the effectiveness and potential of our approach in improving the efficiency of data processing and machine learning workloads on modern hardware.

Long Short Term Memory Based Traffic Prediction Using Multi-Source Data

AbstractTraffic prediction is a task where the goal is to determine the number and type of vehicles, or some other traffic related metric, at certain time point. In addition to predicting the short-term evolution of traffic, prediction can be done for estimating traffic for distant future based on the trends found in historical traffic data, which is a critical component of traffic simulators being able to spawn realistic number of vehicles under prevailing situation. Such prediction system needs to be dependent on the characteristics of the situation and not the preceding traffic flow. This work presents a deep learning based prediction pipeline that uses a Long Short Term Memory (LSTM) network to map temporal, weather and traffic accident data accurately into traffic flow to predict traffic flow over multiple timesteps from various non-traffic inputs. Traffic data can then be produced based on independent data like weather forecasts and be used for other applications. As far as we know, no previous traffic predictor combines so many input variables to predict traffic flow with vehicle type information. To make the event based traffic accident dataset compatible with time series data, a novel preprocessing step based on power law decay phenomenon is added. Quantitative experiments show that the proposed preprocessing step and optimized hyperparameters improve the accuracy of the predictor on multiple metrics compared to a model without accident information. In two established statistical evaluation metrics, Mean Absolute Error and Mean Squared Error, the improvement was over $$20 \%$$ 20 % for certain vehicle types.

Optimization of high-throughput marker systems for genomic prediction in alfalfa family bulks.

Alfalfa (Medicago sativa L.) is a perennial forage legume esteemed for its exceptional quality and dry matter yield (DMY); however, alfalfa has historically exhibited low genetic gain for DMY. Advances in genotyping platforms paved the way for a cost-effective application of genomic prediction in alfalfa family bulks. In this context, the optimization of marker density holds potential to reallocate resources within genomic prediction pipelines. This study aimed to (i) test two genotyping platforms for population structure discrimination and predictive ability (PA) of genomic prediction models (G-BLUP) for DMY, and (ii) explore optimal levels of marker density to predict DMY in family bulks. For this, 160 nondormant alfalfa families were phenotyped for DMY across 11 harvests and genotyped via targeted sequencing using Capture-seq with 17K probes and the DArTag 3K panel. Both platforms discriminated similarly against the population structure and resulted in comparable PA for DMY. For genotyping optimization, different levels of marker density were randomly extracted from each platform. In both cases, a plateau was achieved around 500 markers, yielding similar PA as the full set of markers. For phenotyping optimization, models with 500 markers built with data from five harvests resulted in similar PA compared to the full set of 11 harvests and full set of markers. Altogether, genotyping and phenotyping efforts were optimized in terms of number of markers and harvests. Capture-seq and DArTag yielded similar results and have the flexibility to adjust their panels to meet breeders' needs in terms of marker density.

Explainable brain age prediction: a comparative evaluation of morphometric and deep learning pipelines

Brain age, a biomarker reflecting brain health relative to chronological age, is increasingly used in neuroimaging to detect early signs of neurodegenerative diseases and support personalized treatment plans. Two primary approaches for brain age prediction have emerged: morphometric feature extraction from MRI scans and deep learning (DL) applied to raw MRI data. However, a systematic comparison of these methods regarding performance, interpretability, and clinical utility has been limited. In this study, we present a comparative evaluation of two pipelines: one using morphometric features from FreeSurfer and the other employing 3D convolutional neural networks (CNNs). Using a multisite neuroimaging dataset, we assessed both model performance and the interpretability of predictions through eXplainable Artificial Intelligence (XAI) methods, applying SHAP to the feature-based pipeline and Grad-CAM and DeepSHAP to the CNN-based pipeline. Our results show comparable performance between the two pipelines in Leave-One-Site-Out (LOSO) validation, achieving state-of-the-art performance on the independent test set (MAE=3.21 with DNN and morphometric features and MAE=3.08 with a DenseNet-121 architecture). SHAP provided the most consistent and interpretable results, while DeepSHAP exhibited greater variability. Further work is needed to assess the clinical utility of Grad-CAM. This study addresses a critical gap by systematically comparing the interpretability of multiple XAI methods across distinct brain age prediction pipelines. Our findings underscore the importance of integrating XAI into clinical practice, offering insights into how XAI outputs vary and their potential utility for clinicians.

Conditional Goal-Oriented Trajectory Prediction for Interacting Vehicles.

Predicting future trajectories of pairwise traffic agents in highly interactive scenarios, such as cut-in, yielding, and merging, is challenging for autonomous driving. The existing works either treat such a problem as a marginal prediction task or perform single-axis factorized joint prediction, where the former strategy produces individual predictions without considering future interaction, while the latter strategy conducts conditional trajectory-oriented prediction via agentwise interaction or achieves conditional rollout-oriented prediction via timewise interaction. In this article, we propose a novel double-axis factorized joint prediction pipeline, namely, conditional goal-oriented trajectory prediction (CGTP) framework, which models future interaction both along the agent and time axes to achieve goal and trajectory interactive prediction. First, a goals-of-interest network (GoINet) is designed to extract fine-grained features of goal candidates via hierarchical vectorized representation. Furthermore, we propose a conditional goal prediction network (CGPNet) to produce multimodal goal pairs in an agentwise conditional manner, along with a newly designed goal interactive loss to better learn the joint distribution of the intermediate interpretable modes. Explicitly guided by the goal-pair predictions, we propose a goal-oriented trajectory rollout network (GTRNet) to predict scene-compliant trajectory pairs via timewise interactive rollouts. Extensive experimental results confirm that the proposed CGTP outperforms the state-of-the-art (SOTA) prediction models on the Waymo open motion dataset (WOMD), Argoverse motion forecasting dataset, and In-house cut-in dataset. Code is available at https://github.com/LiDinga/CGTP/.

Based on computer simulation and experimental verification: mining and characterizing novel antimicrobial peptides from soil microbiome

Antimicrobial peptides (AMPs) show great promise for enhancing food safety and extending shelf life, but traditional screening methods are complex and costly. To address these issues, we developed a deep learning-based prediction pipeline to identify potential AMPs from soil metagenomic data, achieving high accuracy (92.71 %) and precision (91.29 %). Based on model scoring, surface charge, and Hemopred and ToxinPred screenings, we identified nine candidate peptides. Peptide P4 (GTAWRWHYRARS) showed the best binding affinity to MrkH in molecular docking studies and was validated through molecular dynamics simulations. The chemically synthesized P4 demonstrated significant antimicrobial activity against Klebsiella pneumoniae, Escherichia coli, and Staphylococcus aureus, indicating its potential as an effective alternative to traditional food antimicrobial agents. This study highlights the effectiveness of our integrated prediction pipeline for discovering new AMPs.

MLR-predictor: a versatile and efficient computational framework for multi-label requirements classification.

Requirements classification is an essential task for development of a successful software by incorporating all relevant aspects of users' needs. Additionally, it aids in the identification of project failure risks and facilitates to achieve project milestones in more comprehensive way. Several machine learning predictors are developed for binary or multi-class requirements classification. However, a few predictors are designed for multi-label classification and they are not practically useful due to less predictive performance. MLR-Predictor makes use of innovative OkapiBM25 model to transforms requirements text into statistical vectors by computing words informative patterns. Moreover, predictor transforms multi-label requirements classification data into multi-class classification problem and utilize logistic regression classifier for categorization of requirements. The performance of the proposed predictor is evaluated and compared with 123 machine learning and 9 deep learning-based predictive pipelines across three public benchmark requirements classification datasets using eight different evaluation measures. The large-scale experimental results demonstrate that proposed MLR-Predictor outperforms 123 adopted machine learning and 9 deep learning predictive pipelines, as well as the state-of-the-art requirements classification predictor. Specifically, in comparison to state-of-the-art predictor, it achieves a 13% improvement in macro F1-measure on the PROMISE dataset, a 1% improvement on the EHR-binary dataset, and a 2.5% improvement on the EHR-multiclass dataset. As a case study, the generalizability of proposed predictor is evaluated on softwares customer reviews classification data. In this context, the proposed predictor outperformed the state-of-the-art BERT language model by F-1 score of 1.4%. These findings underscore the robustness and effectiveness of the proposed MLR-Predictor in various contexts, establishing its utility as a promising solution for requirements classification task.

A real-time predicting online tool for detection of people’s emotions from Arabic tweets based on big data platforms

Emotion prediction is a subset of sentiment analysis that aims to extract emotions from text, speech, or images. The researchers posit that emotions determine human behavior, making the development of a method to recognize emotions automatically crucial for use during global crises, such as the COVID-19 pandemic. In this paper, a real-time system is developed that identifies and predicts emotions conveyed by users in Arabic tweets regarding COVID-19 into standard six emotions based on the big data platform, Apache Spark. The system consists of two main stages: (1) Developing an offline model and (2) Online emotion prediction pipeline. For the first stage, two different approaches: The deep Learning (DL) approach and the Transfer Learning-based (TL) approach to find the optimal classifier for identifying and predicting emotion. For DL, three classifiers are applied: Convolutional Neural Network (CNN), Gated Recurrent Unit (GRU), and Bidirectional GRU (BiGRU). For TL, five models are applied: AraBERT, ArabicBERT, ARBERT, MARBERT, and QARiB. For the second stage, create a Transmission Control Protocol (TCP) socket between Twitter’s API and Spark used to receive streaming tweets and Apache Spark to predict the label of tweets in real-time. The experimental results show that the QARiB model achieved the highest Jaccard accuracy (65.73%), multi-accuracy (78.71%), precision-micro (78.71%), recall-micro (78.71%), f-micro (78.71%), and f-macro (78.55%). The system is available as a web-based application that aims to provide a real-time visualization of people’s emotions during a crisis.

Accurate RNA 3D structure prediction using a language model-based deep learning approach

Accurate prediction of RNA three-dimensional (3D) structures remains an unsolved challenge. Determining RNA 3D structures is crucial for understanding their functions and informing RNA-targeting drug development and synthetic biology design. The structural flexibility of RNA, which leads to the scarcity of experimentally determined data, complicates computational prediction efforts. Here we present RhoFold+, an RNA language model-based deep learning method that accurately predicts 3D structures of single-chain RNAs from sequences. By integrating an RNA language model pretrained on ~23.7 million RNA sequences and leveraging techniques to address data scarcity, RhoFold+ offers a fully automated end-to-end pipeline for RNA 3D structure prediction. Retrospective evaluations on RNA-Puzzles and CASP15 natural RNA targets demonstrate the superiority of RhoFold+ over existing methods, including human expert groups. Its efficacy and generalizability are further validated through cross-family and cross-type assessments, as well as time-censored benchmarks. Additionally, RhoFold+ predicts RNA secondary structures and interhelical angles, providing empirically verifiable features that broaden its applicability to RNA structure and function studies.

CTIM-20. IDENTIFYING CANDIDATE ANTIGEN TARGETS FOR PEDIATRIC BRAIN TUMOR IMMUNOTHERAPIES THROUGH COMPREHENSIVE IMMUNOGENOMIC ANALYSIS

Abstract BACKGROUND Identifying tumor rejection antigens continues to be a significant challenge in the development of effective immunotherapies and their application to pediatric brain tumors. METHODS To identify candidate antigen targets for Medulloblastoma (primary MB, n=170; recurrent MB, n=21), Diffuse Intrinsic Pontine Glioma (DIPG, n=10), and Pediatric High-Grade Astrocytoma (HGA, n=12) for adoptive cellular therapy, we conducted a comprehensive immunogenomic analysis of the transcription profiles of the most aggressive pediatric brain tumors and performed in silico antigen prediction across a wide range of antigen classes. IMPORTANCE OF THE STUDY Pediatric brain tumors are heterogeneous diseases, and current standard treatments do not adequately address this variability. Newer and safer patient-specific treatment approaches are needed for high-risk patients who do not respond to standard therapies. Cellular immunotherapy could be crucial for improving survival and reducing morbidity. For an effective immune response against these brain tumors, appropriate tumor antigens must be targeted. While subgroup-specific genetic mutations in medulloblastoma patients have been reported, the immunogenicity of these genetic alterations remains unknown. Using a custom antigen prediction pipeline, we identified potential tumor rejection antigens with significant implications for the development of cellular therapies for MB, DIPG, and HGA. RESULTS Antigen prediction analysis revealed that most patients express few candidate neoantigen targets that pass all filtering criteria. Overall, the proportion of predicted tumor-associated antigens (TAAs) was higher than that of neoantigens and gene fusions for all molecular subtypes, except for SHH, which exhibited a higher neoantigen burden. Notably, cancer-testis antigens and previously unrecognized neurodevelopmental antigens were found to be expressed in most patients across all medulloblastoma subgroups. Although the absolute immune cell content is predicted to be low, immune gene-signature analysis revealed subgroup-specific differences.

Finding representative group fairness metrics using correlation estimations

It is of critical importance to be aware of the historical discrimination embedded in the data and to consider a fairness measure to reduce bias throughout the predictive modeling pipeline. Given various notions of fairness defined in the literature, investigating the correlation and interaction among metrics is vital for addressing unfairness. Practitioners and data scientists should be able to comprehend each metric and examine their impact on one another given the context, use case, and regulations. Exploring the combinatorial space of different metrics for such examination is burdensome. To alleviate the burden of selecting fairness notions for consideration, we propose a framework that estimates the correlation among fairness notions. Our framework consequently identifies a set of diverse and semantically distinct metrics as representative of a given context. We propose a Monte Carlo sampling technique for computing the correlations between fairness metrics by indirect and efficient perturbation in the model space. Using the estimated correlations, we then find a subset of representative metrics. The paper proposes a generic method that can be generalized to any arbitrary set of fairness metrics. We showcase the validity of the proposal using comprehensive experiments on real-world benchmark datasets.

Prediction of femoral head collapse in osteonecrosis using deep learning segmentation and radiomics texture analysis of MRI

BackgroundFemoral head collapse is a critical pathological change and is regarded as turning point in disease progression in osteonecrosis of the femoral head (ONFH). In this study, we aim to build an automatic femoral head collapse prediction pipeline for ONFH based on magnetic resonance imaging (MRI) radiomics.MethodsIn the segmentation model development dataset, T1-weighted MRI of 222 hips from two hospitals were retrospectively collected and randomly split into training (n = 190) and test (n = 32) sets. In the prognosis prediction model development dataset, 206 hips were also retrospectively collected from two hospitals and divided into training set (n = 155) and external test set (n = 51) according to data source. A deep learning model for automatic lesion segmentation was trained with nnU-Net, from which three-dimensional regions of interest were segmented and a total of 107 radiomics features were extracted. After intra-class correlation coefficients screening, feature correlation coefficient screening and Least Absolute Shrinkage and Selection Operator regression feature selection, a machine learning model for ONFH prognosis prediction was trained with Logistic Regression (LR) and Light Gradient Boosting Machine (LightGBM) algorithm.ResultsThe segmentation model achieved an average dice similarity coefficient of 0.848 and an average 95% Hausdorff distance of 3.794 in the test set, compared to the manual segmentation results. After feature selection, nine radiomics features were included in the prognosis prediction model. External test showed that the LightGBM model exhibited acceptable predictive performance. The area under the curve (AUC) of the prediction model was 0.851 (95% CI: 0.7268–0.9752), with an accuracy of 0.765, sensitivity of 0.833, and specificity of 0.727. Decision curve analysis showed that the LightGBM model exhibited favorable clinical utility.ConclusionThis study presents an automated pipeline for predicting femoral head collapse in ONFH with acceptable performance. Further research is necessary to determine the clinical applicability of this radiomics-based approach and to assess its potential to assist in treatment decision-making for ONFH.

Novel programming pipeline for herbicide resistance prediction

Novel programming pipeline for herbicide resistance prediction

3D CNN for neuropsychiatry: Predicting Autism with interpretable Deep Learning applied to minimally preprocessed structural MRI data.

Predictive modeling approaches are enabling progress toward robust and reproducible brain-based markers of neuropsychiatric conditions by leveraging the power of multivariate analyses of large datasets. While deep learning (DL) offers another promising avenue to further advance progress, there are challenges related to implementation in 3D (best for MRI) and interpretability. Here, we address these challenges and describe an interpretable predictive pipeline for inferring Autism diagnosis using 3D DL applied to minimally processed structural MRI scans. We trained 3D DL models to predict Autism diagnosis using the openly available ABIDE I and II datasets (n = 1329, split into training, validation, and test sets). Importantly, we did not perform transformation to template space, to reduce bias and maximize sensitivity to structural alterations associated with Autism. Our models attained predictive accuracies equivalent to those of previous machine learning (ML) studies, while side-stepping the time- and resource-demanding requirement to first normalize data to a template. Our interpretation step, which identified brain regions that contributed most to accurate inference, revealed regional Autism-related alterations that were highly consistent with the literature, encompassing a left-lateralized network of regions supporting language processing. We have openly shared our code and models to enable further progress towards remaining challenges, such as the clinical heterogeneity of Autism and site effects, and to enable the extension of our method to other neuropsychiatric conditions.

AI-Based Smart Proxy Models for Accurate Oil Rate Prediction and Efficient Pipeline Monitoring

This research develops an advanced AI-based smart proxy model to significantly enhance the prediction of oil rates and the monitoring of crucial operational parameters such as temperature and pressure in oil field pipeline management. By integrating real-time data from Multiphase Flow Meters (MPFM) with sophisticated simulation outputs, the study introduces a dual-model approach that overcomes the limitations of traditional methods, improving both efficiency and cost-effectiveness. Model 1 employs high-precision real-time MPFM data to provide accurate oil rate predictions. By focusing on critical control points within expansive pipeline networks, this model strategically reduces dependency on extensive MPFM deployment, achieving substantial cost reductions while maintaining rigorous measurement standards. The incorporation of real-time data ensures immediate responsiveness to operational changes, facilitating accurate and reliable insights essential for effective pipeline management. Model 2 utilizes an AI-driven smart proxy to refine the outputs of conventional flow simulators such as OLGA. This model addresses computational challenges including high runtime and numerical convergence issues by selecting the most reliable and accurate simulation outputs. It provides rapid and dependable insights into flow dynamics, supporting timely operational decisions and proactive management that enhance the safety and efficiency of pipeline networks. The integration of Model 1 and Model 2 ensures localized precision and extends analytical capabilities across the entire pipeline network, significantly enhancing predictive accuracy. This harmonized approach not only sets new standards for flow assurance and pipeline management but also illustrates the transformative impact of AI on operational strategies in the hydrocarbon sector.

Unmasking AlphaFold to integrate experiments and predictions in multimeric complexes.

Since the release of AlphaFold, researchers have actively refined its predictions and attempted to integrate it into existing pipelines for determining protein structures. These efforts have introduced a number of functionalities and optimisations at the latest Critical Assessment of protein Structure Prediction edition (CASP15), resulting in a marked improvement in the prediction of multimeric protein structures. However, AlphaFold's capability of predicting large protein complexes is still limited and integrating experimental data in the prediction pipeline is not straightforward. In this study, we introduce AF_unmasked to overcome these limitations. Our results demonstrate that AF_unmasked can integrate experimental information to build larger or hard to predict protein assemblies with high confidence. The resulting predictions can help interpret and augment experimental data. This approach generates high quality (DockQ score > 0.8) structures even when little to no evolutionary information is available and imperfect experimental structures are used as a starting point. AF_unmasked is developed and optimised to fill incomplete experimental structures (structural inpainting), which may provide insights into protein dynamics. In summary, AF_unmasked provides an easy-to-use method that efficiently integrates experiments to predict large protein complexes more confidently.

OriTDB: a database of the origin-of-transfer regions of bacterial mobile genetic elements.

Conjugation and mobilization are two important pathways of horizontal transfer of bacterial mobile genetic elements (MGEs). The origin-of-transfer (oriT) region is crucial for this process, serving as a recognition site for relaxase and containing the DNA nicking site (nic site), which initiates the conjugation or mobilization. Here, we present a database of the origin-of-transfer regions of bacterial MGEs, oriTDB (https://bioinfo-mml.sjtu.edu.cn/oriTDB2/). Incorporating data from text mining and genome analysis, oriTDB comprises 122 experimentally validated and 22927 predicted oriTs within bacterial plasmids, Integrative and Conjugative Elements, and Integrative and Mobilizable Elements. Additionally, oriTDB includes details about associated relaxases, auxiliary proteins, type IV coupling proteins, and a gene cluster encoding the type IV secretion system. The database also provides predicted secondary structures of oriT sequences, dissects oriT regions into pairs of inverted repeats, nic sites, and their flanking conserved sequences, and offers an interactive visual representation. Furthermore, oriTDB includes an enhanced oriT prediction pipeline, oriTfinder2, which integrates a functional annotation module for cargo genes in bacterial MGEs. This resource is intended to support research on bacterial conjugative or mobilizable elements and promote an understanding of their cargo gene functions.

NeuSyRE: Neuro-symbolic visual understanding and reasoning framework based on scene graph enrichment

Exploring the potential of neuro-symbolic hybrid approaches offers promising avenues for seamless high-level understanding and reasoning about visual scenes. Scene Graph Generation (SGG) is a symbolic image representation approach based on deep neural networks (DNN) that involves predicting objects, their attributes, and pairwise visual relationships in images to create scene graphs, which are utilized in downstream visual reasoning. The crowdsourced training datasets used in SGG are highly imbalanced, which results in biased SGG results. The vast number of possible triplets makes it challenging to collect sufficient training samples for every visual concept or relationship. To address these challenges, we propose augmenting the typical data-driven SGG approach with common sense knowledge to enhance the expressiveness and autonomy of visual understanding and reasoning. We present a loosely-coupled neuro-symbolic visual understanding and reasoning framework that employs a DNN-based pipeline for object detection and multi-modal pairwise relationship prediction for scene graph generation and leverages common sense knowledge in heterogenous knowledge graphs to enrich scene graphs for improved downstream reasoning. A comprehensive evaluation is performed on multiple standard datasets, including Visual Genome and Microsoft COCO, in which the proposed approach outperformed the state-of-the-art SGG methods in terms of relationship recall scores, i.e. Recall@K and mean Recall@K, as well as the state-of-the-art scene graph-based image captioning methods in terms of SPICE and CIDEr scores with comparable BLEU, ROGUE and METEOR scores. As a result of enrichment, the qualitative results showed improved expressiveness of scene graphs, resulting in more intuitive and meaningful caption generation using scene graphs. Our results validate the effectiveness of enriching scene graphs with common sense knowledge using heterogeneous knowledge graphs. This work provides a baseline for future research in knowledge-enhanced visual understanding and reasoning. The source code is available at https://github.com/jaleedkhan/neusire.

Leveraging neural networks to correct FoldX free energy estimates.

Proteins play a pivotal role in many biological processes, and changes in their amino acid sequences can lead to dysfunction and disease. These changes can affect protein folding or interaction with other biomolecules, such as preventing antibodies from inhibiting a viral infection or causing proteins to misfold. The ability to predict the effects of mutations in proteins is crucial. Although experimental techniques can accurately quantify the effect of mutations on protein folding free energies and protein-protein binding free energies, they are often time-consuming and costly. By contrast, computational techniques offer fast and cost-effective alternatives for estimating free energies, but they typically suffer from lower accuracy. Enhancing the accuracy of computational predictions is therefore of high importance, with the potential to greatly impact fields ranging from drug design to understanding disease mechanisms. One such widely used computational method, FoldX, is capable of rapidly predicting the relative folding stability ( ) for a protein as well as the relative binding affinity ( ) between proteins using a single protein structure as input. However, it can suffer from low accuracy, especially for antibody-antigen systems. In this work, we trained a neural network on FoldX output to enhance its prediction accuracy. We first performed FoldX calculations on the largest datasets available for mutations that affect binding (SKEMPIv2) and folding (ProTherm4) with experimentally measured . Features were then extracted from the FoldX output files including its prediction for . We then developed and optimized a neural network framework to predict the difference between FoldX's estimated and the experimental data, creating a model capable of producing a correction factor. Our approach showed significant improvements in Pearson correlation performance. For single mutations affecting folding, the correlation improved from a baseline of 0.3 to 0.66. In terms of binding, performance increased from 0.37 to 0.61 for single mutations and from 0.52 to 0.81 for double mutations. For epistasis, the correlation for binding affinity (both singles and doubles) improved from 0.19 to 0.59. Our results also indicated that models trained on double mutations enhanced accuracy when predicting higher-order mutations (such as triple or quadruple mutations), whereas models trained on singles did not. This suggests that interaction energy and epistasis effects present in the FoldX output are not fully utilized by FoldX itself. Once trained, these models add minimal computational time but provide a substantial increase in performance, especially for higher-order mutations and epistasis. This makes them a valuable addition to any free energy prediction pipeline using FoldX. Furthermore, we believe this technique can be further optimized and tested for predicting antibody escape, aiding in the efficient development of watch lists.

Enhancing Deep Learning-Based City-Wide Traffic Prediction Pipelines Through Complexity Analysis

Deep learning models can effectively capture the non-linear spatiotemporal dynamics of city-wide traffic forecasting. Evidence of varying deep learning model performance between different cities, different prediction horizons, different scales, specific city regions, and during particular hours of the day abounds in the literature on deep learning-based traffic prediction, yet a unified metric to quantify the complexity of different prediction tasks does not exist. This paper proposes two metrics—model complexity (MC) and intrinsic complexity (IC). While MC quantifies the effective complexity of deep learning models for city-wide traffic prediction tasks, the IC quantifies the underlying complexity of the prediction task. Being an effective complexity metric, MC depends on the model and the data. The IC depends only on the data and is invariant to the model being used. Both metrics are validated through systematic experimentation using traffic volume data from three cities. Finally, we demonstrate how these metrics can improve the workflows for deep learning-based data-driven traffic prediction pipelines and deployment by reducing the hyperparameter search scope and comparing the effectiveness of different design pathways.