Computational and Structural Biotechnology Journal

Oncogenic mutations are known to occur at non-random rates and specific hotspots, yet the structural and thermodynamic factors underlying these hotspots remain poorly understood. This study presents molecular dynamics simulations of 22 cancer hotspots across 10 oncogene families, simulated as histone-DNA complexes. Across 18 of 22 mutations, thermodynamic effects were predominantly localized to within 6 [A] of the DNA-histone interface (mean capture 104%), with van der Waals interactions driving the effect at the thermodynamic extremes, indicating that oncogenic mutations alter nucleosome binding through precise local contact changes rather than global structural rearrangements. Further, analysis revealed a gene-family correlated pattern in thermodynamic stability. RAS family mutations showed a consistent trend toward nucleosome stabilization (mean {Delta}{Delta}G = -4.82 kcal mol-1), while kinase domain mutations trended toward destabilization (mean {Delta}{Delta}G = +40.12 kcal mol-1), a difference reaching statistical significance in this exploratory analysis (Mann-Whitney U, p = 0.008). This interface-specific mechanism, combined with the gene-family-correlated thermodynamic pattern, provides a biophysical framework for understanding nucleosome-level contributions to cancer hotspot mutation biology.

The perception of flavor during food consumption is a complex process governed by the dynamic release of aroma compounds, which is influenced by the interaction between the food matrix and human physiology. While mechanistic models have been developed to simulate these phenomena, their practical application has often been limited by the lack of flexible software tools capable of integrating diverse food structures and real-time physiological data. This study addresses this gap by incorporating in vivo mechanistic models for both liquid foods and melting solid matrices in a modular Python-based framework: digimouth. The modeling approach treats the oral, pharyngeal, and nasal cavities as a series of interconnected reactors with mass transfer governed by ordinary differential equations (ODEs). These models explicitly integrate key physiological parameters, including salivary flow, bidirectional respiratory airflow, and discrete events such as mastication and swallowing. For solid foods, the framework incorporates additional mechanisms for matrix dissolution, temperature-dependent melting, and dynamic surface area change. The models were validated using experimental data from 30 subjects who consumed aroma solutions and flavored gelatin gummy discs while monitored by PTR-MS (Proton Transfer Reaction - Mass Spectrometry). The simulations successfully reproduced the main kinetic features and release profiles observed experimentally for both types of matrices. Despite some limitations related to the self-reporting of physiological events and the non-inclusion of initial orthonasal mass transfer in the models, the developed software package proved to be a robust and evolutive tool. It offers a foundation for exploring inter-individual variability and the impact of complex food structures on sensory perception.

The modeling of aroma release in food science is a complex, multidisciplinary field integrating mass transfer theory and principles of physical chemistry. While mechanistic models have long been used to understand volatile release, their practical adoption has been hindered because existing implementations are often private, "in-house" tools that lack reproducibility and extensibility. This paper introduces digimouth, an open-source Python framework designed to provide a unified, modular, and reproducible environment for simulating and fitting aroma release kinetics. A key innovation of digimouth is its handling of events and stages. Events represent discrete, instantaneous changes in the system (such as opening a bottle or swallowing), while stages allow for multi-phase simulations where experimental conditions or physical dynamics change over time. To demonstrate the tool's utility, the authors implemented established in vitro convection models and performed sensitivity analyses on parameters such as air-water partition coefficients and mass transfer coefficients. The package also includes an optimization module based on the least-squares method to fit parameters to experimental data. Validation against real-time data from Proton Transfer Reaction-Time of Flight- Mass Spectrometry (PTR-ToF-MS) shows a satisfactory fit, with an average error of approximately 2%. Ultimately, digimouth facilitates robust model development and hypothesis testing, offering the food science community a collaborative tool to better understand the mechanisms of flavor release.

T cell receptor (TCR) specificity prediction is critical for understanding adaptive immunity and for enabling the development of therapeutic interventions. Current machine learning models for predicting TCR-peptide-MHC interactions perform poorly, despite large efforts by many groups and advances in protein language models. This work identifies and addresses the fundamental issues that might be hindering progress in this field, focusing on current approaches for constructing training and validation sets, particularly on the strategies used by commonly used methods, which generate negative data either by shuffling TCRs between epitopes or by pairing epitopes with TCRs from background datasets of healthy donors. Here, we visualize and quantify the overlap between positive and generated negative TCR sets produced by both strategies using a range of sequence embedding approaches, including simple encodings (one-hot and BLOSUM62), general protein language models (ESM-2 and ProtGPT2), and a TCR-specific embedding model (TCR-BERT). Using paired {beta}TCR data from VDJdb for two well-studied epitopes (GILGFVFTL and KLGGALQAK), we train Random Forest classifiers on each embedding representation and evaluate their ability to distinguish positive from negative data examples. We find that general sequence embedding methods struggle to differentiate between positive and negative TCRs. We further show that the choice of negative data generation method significantly impacts classifier performance. Our findings highlight fundamental limitations in current TCR specificity training data and general protein language models, and underscore the need for TCR-specific embeddings, experimentally validated negative datasets, transparent reporting of negative data construction, and epitope-specific modeling approaches.

Abstract Background: Immune-mediated skin diseases frequently co-occur, but the molecular architecture connecting clinically distinct disorders, and the extent to which molecular proximity predicts shared therapeutic opportunity, remain incompletely defined. We constructed a multilayer skin diseasome integrating transcriptomic overlap, pathway enrichment, protein-interactome topology, curated disease associations, real-world clinical co-occurrence, and drug-perturbation signatures across psoriasis (PS), atopic dermatitis (AD), hidradenitis suppurativa (HS), vitiligo (VL), and chronic spontaneous urticaria (CSU), using sodium lauryl sulfate-induced irritant contact dermatitis (SLS-ICD) as a mechanistically distinct barrier-injury comparator. Methods: Differentially expressed genes (DEGs) were derived from publicly available NCBI Gene Expression Omnibus datasets and compared using exclusive pairwise and higher-order intersections. Gene Ontology biological-process enrichment characterized shared and condition-specific programs. Interactome-based network separation (S_AB) quantified the relative topological proximity of disease modules. DisGeNET enrichment generated candidate disease +associations, a selected subset of which was examined descriptively for patient co-occurrence in TriNetX. Candidate compounds were ranked by reversal of disease-associated expression signatures against LINCS L1000 perturbation profiles. Results: The largest exclusive DEG intersection occurred between PS and HS (2,060 genes), followed by AD-PS-HS (490), VL-PS (464), AD-PS (402), and VL-PS-HS (379). Fourteen genes were shared across PS, AD, HS, VL, and CSU but not SLS-ICD. Neutrophil-associated processes recurred across the immune-mediated conditions, whereas SLS-ICD was dominated by keratinization and epidermal-differentiation programs. PS and HS had the lowest observed network-separation value (S_AB = -0.594); SLS-ICD had positive S_AB values in all five comparisons, although this separation did not persist once module sizes were harmonized. DisGeNET linked the index conditions to diverse immune, cardiovascular, neurologic, and neoplastic categories; 42 of these candidate diseases were observed to co-occur with an index condition in the queried TriNetX cohorts. Drug-signature overlap was greatest between PS and CSU, despite their marginal interactome separation, indicating that interactome proximity and predicted pharmacologic reversal capture different dimensions of disease relatedness. mTOR/PI3K-pathway inhibitors were prominent among the psoriasis reversal candidates, while methotrexate and mycophenolate mofetil emerged from both the CSU signature and the curated gene set associated with HIV infection. Conclusion: Multilayer integration identified a connected molecular core among five immune-mediated skin diseases and a distinct SLS-ICD injury module. The discordance between interactome proximity and drug-signature overlap suggests that no single layer adequately defines disease relatedness. These findings provide a hypothesis-generating framework for prioritizing disease links and repurposing candidates, but controlled epidemiologic analyses and experimental validation are required before clinical or therapeutic inference.

Chromatin profiling methods, such as ChIP-seq (chromatin immunoprecipitation followed by sequencing), are used to characterize the genomic localization of DNA-associated proteins. While ChIP and ChIP-seq have been used for decades, an orthogonal approach, CUT&Tag (Cleavage Under Targets and Tagmentation), is gaining popularity as an efficient and cost-effective alternative. Although previous comparative studies note discrepancies in signal-to-noise ratios and detection bias at certain genomic regions, many differences between ChIP-seq and CUT&Tag results remain largely unreconciled. Here, we systematically investigate the disagreeing signals captured by these two methods across well annotated genomic regions. We assess multiple histone mark profiles generated by different groups in two cell lines (K562 and MCF-7). Overall, our analysis indicates that compared to ChIP-seq, CUT&Tag may have limited sensitivity in low-GC environments and, as previously observed, increased signal in hyper-accessible chromatin. Notably, within GC-poor regions of active gene bodies and Polycomb-repressed domains, CUT&Tag exhibits a loss of H3K36me3 and H3K27me3 signal, respectively, where occupancy of these histone marks is otherwise expected. Further, active promoters with discrepant H3K4me3 and H3K27ac signal between the two assays differ systematically in GC content and chromatin accessibility. Promoters differentially enriched for CUT&Tag signal relative to ChIP-seq generally show higher, broader GC-content profiles and higher DNase-seq and ATAC-seq signal, while promoters enriched for ChIP-seq signal harbor the opposite characteristics. A local bias for high GC content and/or chromatin accessibility in CUT&Tag may also explain its differing signal patterns at nucleosome-depleted regions (NDRs) compared to ChIP-seq and MNase-seq promoter profiles. Altogether, our results highlight that studies focused on profiling the intensity and structure of histone modification occupancy can be sensitive to potential biases of CUT&Tag to GC content and chromatin accessibility. These characteristics should be accounted for when selecting a chromatin profiling approach, analyzing and interpreting data as well as drawing biological conclusions.

Predicting implantation outcomes in vitro fertilization (IVF) is challenging because clinically relevant information becomes available at different points during treatment. In this study, we developed a stage-aware, leakage-conscious, and explainable machine-learning framework to predict implantation using a public IVF dataset. {beta}-HCG was excluded to avoid potential downstream target leakage. The data were divided into training, validation, and independent test sets. Several machine-learning models, including Logistic Regression, Random Forest (RF), Extra Trees, XGBoost, and Hist Gradient Boosting, were optimized using stratified cross-validation. A fixed RF model was then evaluated at three stages of the IVF process, using baseline couple characteristics, gamete and fertilization variables, and embryological variables, respectively. Model performance improved as additional treatment-related information became available, with the largest improvement observed after the inclusion of embryological variables. The final RF model showed good predictive performance on the independent test set. The number of embryos was consistently identified as the most important predictor across RF, permutation, and SHAP analyses. Overall, the findings suggest that IVF implantation prediction becomes more accurate as later-stage treatment information is incorporated, particularly information related to embryological characteristics. The proposed approach also provides interpretable predictions while reducing the risk of information leakage. Further evaluation using independent clinical datasets is needed before the framework can be considered for clinical use.

Recent biological research has shown significant interest and progress towards adoption of agentic AI systems. Domain specific agentic platforms with access to large collections of biological tools with code generation and/or querying capabilities, such as Biomni (illustrative case in point), are representative of the capability to orchestrate automated research workflows. Such platforms enable collaborative AI workspace for drug discovery, genomics, bioinformatics, literature research, and experimental planning for biologists and researchers. When such platforms are used for the aforementioned workflows, individually validating every agentic decision and the research process by which it was reached becomes challenging due to the scale of decisions involved and oftentimes, insufficient disclosure of assumptions and methodological choices, limited traceability of evidence, and interpretive judgments embedded throughout the workflow, giving rise to verification debt. Verification debt in agentic environments could result in unexpected failures and consequences in downstream experiments. This study focuses on empirical qualitative characterization of verification debt in agentic (prompt to outcomes) workflows. Biomni was chosen the testing AI agentic platform in this study mainly due to the descriptive traces it provides to track every agentic step. Verification debt is examined through two complementary angles: the individual agentic decision steps that carry claims and evidence, and the underlying research process by which the agent arrives at and reports them. This study attempts to characterize verification debt with 10 test cases (covering proteomics, genomics and drug discovery) with varying degree of biological nuance across three identical and independent runs. Verification debt is represented through 9 decision-step metrics and 13 research-process metrics, and also propose the VERIFY framework for evaluating scientific verifiability and validity in agentic workflows.

Quantitative characterization of cell proliferation is central to preclinical drug discovery. Here, we evaluated Random Forest (RF) regression for predicting confluence-based cell growth trends using data from human cancer cell lines and benchmarked its performance against the widely used logistic and Gompertz models. The RF model achieved higher predictive accuracy within the evaluated dataset, particularly at higher initial seeding densities. We further extended the RF framework to predict growth curves across seeding densities. Collectively, these findings highlight the potential of data-driven modeling to complement conventional mathematical approaches and support more informed cell-culture design.

Structure-guided mutation design requires relating a proposed substitution to its molecular environment to anticipate its effects. Existing evaluations of large language models (LLMs) in protein engineering emphasize sequence-based prediction or tool-integrated workflows, leaving unclear whether LLMs themselves can infer mutation effects from structural evidence without dedicated predictors. We evaluate physical intuition-qualitative reasoning that connects structural observations with molecular interaction principles-using a benchmark that tests the same substitution in the same protein across different binding partners, with predictions based on local atomic coordinates and no dedicated mutation-effect predictors. In the tested comparisons, the best-performing LLM achieved group accuracy comparable to the evaluated energy predictors and Biomni, an LLM-driven tool workflow, while the methods differed in their prediction patterns. Most evaluated LLMs achieved higher group accuracy for larger affinity improvements. Structural interventions support selective dependence on mutation-site and partner information, while assessable structural claims largely agree with the supplied coordinates. These findings support direct LLM structural reasoning as a basis for developing rational mutation-design tools that connect molecular observations with partner-specific effects.

Peptides containing noncanonical amino acids (ncAAs) are becoming increasingly important in drug discovery, but their molecular dynamics (MD) simulation remains limited by labor-intensive residue parameterization and topology preparation. Here, we present TopFor (Topology and Residue Force-Field Parameter Preparation), an open-source, peptide-oriented workflow that automates the generation of ncAA residue templates and force-field parameter files for MD simulations with AMBER using established charge-derivation methods. TopFor supports single-residue, whole-peptide, and batch parameterization workflows. The workflow was validated using the MDM2-p53 complex by comparing native AMBER residue definitions with residues processed by TopFor using three charge-derivation methods, each evaluated in three independent 100 ns MD simulations. TopFor-generated residue definitions showed structural and conformational behavior comparable to the native AMBER representation. Applicability to more complex peptide architectures was demonstrated using the head-to-tail cyclic peptide cyclosporin A and the cyclic depsipeptide cyclohexylgriselimycin. At library scale, 1,138 of 1,214 cyclohexylgriselimycin derivatives (93.7%) successfully completed TopFor parameterization and the subsequent initial AMBER-based computational workflow across two derivatization campaigns. TopFor was further integrated into a structure-based peptide discovery workflow based on the DnaN-griselimycin complex. Six GM-derived peptides were computationally prioritized and experimentally evaluated, of which four produced SPR responses against Mycobacterium smegmatis (M. smegmatis) DnaN and three showed concentration-dependent responses. TopFor reduces the manual effort required to prepare ncAA-containing peptides for MD simulations and enables their integration into both library-scale computational screening and structure-based peptide discovery workflows. Keywords: noncanonical amino acids; peptide parameterization; molecular dynamics; AMBER; AmberTools; peptide design; high-throughput virtual screening

Over the past decade, aptamers have emerged as promising therapeutics, with cancer therapeutics as a major research area. Existing computational methods, however, either predict aptamer-target pairs or design aptamers against a specific target. In this study, we present AntiCapt, a computational method for identifying single-stranded DNA (ssDNA) anticancer aptamers (ACAs) using sequence descriptors and fine-tuned nucleotide language models. The main dataset comprises 1,021 experimentally validated ACAs and an equal number of non-aptamer sequences. Comparative analysis revealed distinct patterns in the nucleotide, dinucleotide, and trinucleotide compositions associated with ACAs. We developed machine learning (ML) models using composition, autocorrelation, binary profiles and structural features. Among these, the best-performing composition-based model achieved an AUC of 0.93 on an independent dataset, while chemical descriptor-and structural feature-based models achieved a maximum AUC of 0.89. ML models using pretrained and fine-tuned NLM embeddings were also developed, with fine-tuned HyenaDNA embeddings achieving the highest performance, with an independent AUC of 0.94. Furthermore, we developed a model for discriminating anticancer aptamers from general aptamers. The best-performing models were integrated into AntiCapt, a web server and a standalone tool for predicting, designing, and genome-scale scanning anticancer aptamers (https://webs.iiitd.edu.in/raghava/anticapt/).

Background: The spatial organization of the human genome emerges from an enzyme-guided self-assembly process in which chromatin has a bidirectional relationship with transcription. Central to this architecture are chromatin packing domains, 50-200 nm structures exhibiting mass-scaling behavior. Nascent domains arise through transcription and cohesin/CTCF-mediated loop extrusion, while their maturation depends on conditions that permit differential access of epigenetic enzymes to their dense cores and continuously lower density periphery, resulting in the accumulation of heterochromatic marks in domain cores and euchromatic marks outside of the domain cores. In turn, packing domains support transcription by creating an "ideal zone" peripherally to their cores due to a nonmonotonic dependence of transcription rate on local chromatin density. Results: We investigated the role of heterochromatin protein 1 alpha (HP1), a key chromatin crosslinking factor with affinity to H3K9me3 marks, in maintaining packing domain organization. Paradoxically, short-term HP1 degradation resulted in overall downregulation of genes, contrasting with the canonical view that repression of heterochromatin allows chromatin to become more accessible for transcription. While some packing domains persisted, heterochromatic cores of dense packing domains were disrupted. When exposed to hypoxia, HP1-depleted cells exhibited a dampened transcriptional response, showcasing that when domain maturation is disrupted, stress response pathways are not effectively activated. Conclusions: This work highlights HP1 crosslinking as an essential component of domain maturation, as HP1 loss disrupts domain geometry and inhibits efficient transcriptional reprogramming. Our findings suggest an intriguing perspective on the potential relationship between chromatin architecture and gene regulation that could inform chromatin engineering-based treatments.

Computational safety evaluation of phytochemical libraries commonly relies on models trained on experimentally tested compounds and applied to much larger and chemically diverse collections. An important but often overlooked question is whether predicted compounds are adequately represented within the experimental chemical space underlying these models. Here, we examined this question for the Indian Medicinal Plants, Phytochemistry and Therapeutics (IMPPAT) database using experimentally characterized datasets for three anti-target endpoints: hERG blockade, cytochrome P450 (CYP) inhibition, and pregnane X receptor (PXR) induction. After structure standardization, 14,956 unique IMPPAT compounds were compared with 8,877 hERG, 4,905 CYP, and 4,138 PXR reference compounds using consensus nearest-neighbor similarity based on Morgan and atom-pair fingerprints. At a Tanimoto similarity threshold of 0.80, only 0.92%, 0.30%, and 0.46% of IMPPAT compounds had close measured analogs in the hERG, CYP, and PXR datasets, respectively. Relative to the internal nearest-neighbor distributions of the reference datasets, a typical IMPPAT compound lay near the 1st percentile, indicating substantial chemical-space separation. Compounds with close experimental analogs also showed distinct molecular characteristics, with descriptor-based logistic regression yielding cross-validated AUC values of 0.77-0.80. Plant-level enrichment was markedly reduced after excluding widely distributed metabolites and was further restricted when the nearest measured analog was required to have pIC50 or pEC50 [≥] 5. The remaining signals were concentrated in a limited number of taxa, including Rauvolfia, Sinopodophyllum/Dysosma, Cinchona, and Catharanthus roseus. Overall, most IMPPAT phytochemicals lack close structural representation among currently measured anti-target compounds. Computational safety predictions for these molecules should therefore be interpreted as extrapolations and accompanied by explicit measures of structural support. Absence of a close experimental analog should be regarded as a gap in available evidence rather than evidence of safety.

Bioinformatics software plays a central role in biological research, yet many academic projects lack robust testing and continuous integration (CI) practices. This absence can compromise reproducibility, accessibility, and long-term maintainability. In this study, we conduct an empirical analysis of GitHub repositories spanning bioinformatics and other scientific software domains to examine the relationship between CI, software testing, and software quality. Repository discovery was performed using domain-specific search terms via GitHub's Application Programming Interface (API). This was followed by a layered data extraction pipeline, which included a Large Language Model (LLM) validation step, that connected repositories to published works, institutions, authors, funding, and citation counts. By integrating repository activity, community engagement, and backlog health into a composite Repository Quality Index (RQI), we find that the presence of CI and software tests is significantly associated with higher repository quality. Additionally, repositories with higher academic citation counts tend to cluster in higher RQI tiers, suggesting that software quality, which is driven by proper testing and integration, is associated with academic impact and funding status. These results demonstrate that adopting industry-standard development practices, such as CI and software testing, is a tractable path toward more sustainable and impactful academic software and funding opportunities.

Rule-based languages have proven effective for modeling systems of interacting structured entities as typically encountered in chemistry and molecular biology. We present PyKappa, a rule-based modeling package written in Python whose interpreted nature enables interactive simulation and analysis, including by agentic AI. The package seeks to broaden the base of developers by utilizing a widely known programming language and serves as an easy-to-deploy teaching tool. Using PyKappa, we conduct a case study of phase separation.

Background and Objective: Prescribing multiple drugs simultaneously, known as polypharmacy, is increasingly common in managing chronic diseases such as diabetes and cardiovascular conditions. Studies show that roughly 40% of elderly patients are on five or more medications, putting them at significant risk of adverse drug-drug interactions (DDIs) [1]. While computational methods based on Graph Neural Networks (GNNs) have shown strong potential for predicting these interactions from a drugs molecular structure, most existing models are limited in two important ways: they rely on attention mechanisms that cannot adapt to the unique context of each atom in a molecule, and they only predict whether an interaction exists rather than how dangerous it might be. Methods: We propose HMVA-DDI, a two-stage deep learning framework that addresses both shortcomings. In the first stage, our model predicts whether two drugs will interact. If it detects an interaction, the second stage classifies it into one of three clinical severity levels: minor, moderate, or major. To better capture molecular structure, we replace the standard attention mechanism with a more powerful version called Graph Attention Network v2 (GATv2), which considers each atoms specific context when learning its representation. We also introduce a technique to prevent information loss in deeper network layers, and we tailor the input features separately for each prediction task. Results: Evaluated on two standard benchmarks, DrugBank 5.0.3 [2] and DDInter [3], HMVA-DDI achieves an AUROC of 0.9819 and AUPRC of 0.9759 on binary interaction detection, a marked improvement over the baseline. For severity classification, our best model variant (qcSvr4) reaches an overall AUROC of 0.9444, with a particularly large improvement in detecting low-severity (Minor) interactions, a historically difficult class. Conclusion: HMVA-DDI bridges the gap between computational drug research and clinical usefulness by not only predicting interactions accurately but also indicating interaction severity. This two-level system can help healthcare professionals prioritize warnings, reducing the alert fatigue problem where doctors are overwhelmed by too many non-critical notifications.

This work presents DSSNA (Define Spatial Structure of Nucleic Acids) v2026, a free and open-source standalone GROMACS module for the automated analysis of nucleic acid structure based on atomic coordinates. DSSNA reproduces the core functionality of the X3DNA and DSSR approach while extending it to the analysis of molecular dynamics trajectories through integration with the GROMACS software package. The algorithm identifies and classifies structural features, including canonical and noncanonical base pairs, hydrogen-bond networks, stacking interactions, helices, and many other elements of nucleic acid secondary and tertiary structure. Its modular output system enables users to select the structural metrics required for a particular analysis, while the use of neighbor-search algorithms and data-processing capabilities facilitates the analysis of large trajectory datasets. The performance and applicability of DSSNA were demonstrated using a molecular dynamics trajectory of a GUAA tetraloop mutant of the sarcin-ricin domain from Escherichia coli 23S ribosomal RNA (PDB ID: 1MSY). The analysis showed that only a subset of the observed base pairs and stacking interactions remains stable throughout the trajectory. Persistent canonical base pairs formed the structural core responsible for maintaining the secondary-structure profile, whereas most noncanonical base pairs and stacking interactions were transient. The results demonstrate that DSSNA can be used to characterize both static nucleic acid structures and the temporal evolution of structural interactions in molecular dynamics simulations. The work also provides a glossary of terms relevant to the analysis and interpretation of nucleic acid structures.

Accurately predicting cellular responses to drugs remains a challenge with the potential to reduce experimental screening costs and accelerate drug discovery. Current computational approaches represent compounds through chemical structures, which carry little information regarding their activity within biological systems. We show that gene expression responses, measured in a reference cell line, can instead serve as transferable representations of chemical perturbations. The proposed framework, BioPert, uses biological representations with a small neural network to predict transcriptional delta responses in cell lines of different lineages. Across the Tahoe-100M and LINCS L1000 datasets, BioPert outperforms molecular fingerprints and embeddings learned from chemical structure, which provide limited improvements over random controls. BioPert's prediction correlations reach 0.79 on Tahoe-100M, corresponding to an improvement of 0.34 over the next-best representation. On LINCS, performance varies with the experimental reproducibility of the test conditions, highlighting the impact of batch effects. Notably, BioPert considerably surpasses predictions obtained by copying the reference response, demonstrating that it captures context-dependent effects. Further analysis of a C32-cobimetinib case shows pathway-level accuracy of the predictions, including when the reference and target responses differ. These results open a new path for chemical perturbation prediction and could ultimately reduce the burden of phenotypic screening.

Ligand transport through proteins underlies many aspects of molecular function, but its detailed mechanisms remain largely inaccessible to direct observation. Molecular dynamics simulations could close this gap, but are computationally too demanding for practical purposes, including filtering stages of enzyme design pipelines. Simplified docking-based approaches might be informative, but sometimes the oversimplifications required for high throughput applications might lead to unreliable predictions. Here we present RosettaLigand PathFinder (RLPF), a fast and practical tool explicitly built to ensure conformational continuity of the ligand trajectory, protein and ligand flexibility, and robustness of every run. Built using a flexible RosettaScripts XML interface, it enables fine-grained control over all simulation parameters and adaptability for a wide range of use cases. Applied to several representative systems, RLPF reproduced established transport features and mutation effects, highlighting its ability to produce biologically relevant insights and serve as a filter in enzyme design campaigns. Wide customizability and well-defined scope of RLPF make it a useful addition to the family of tools to study the dynamics of interactions between small molecules and proteins.