Computational Workflow Research Articles

Raman-based machine-learning platform reveals unique metabolic differences between IDHmut and IDHwt glioma.

Formalin-fixed, paraffin-embedded (FFPE) tissue slides are routinely used in cancer diagnosis, clinical decision-making, and stored in biobanks, but their utilization in Raman spectroscopy-based studies has been limited due to the background coming from embedding media. Spontaneous Raman spectroscopy was used for molecular fingerprinting of FFPE tissue from 46 patient samples with known methylation subtypes. Spectra were used to construct tumor/non-tumor, IDH1WT/IDH1mut, and methylation-subtype classifiers. Support vector machine and random forest were used to identify the most discriminatory Raman frequencies. Stimulated Raman spectroscopy was used to validate the frequencies identified. Mass spectrometry of glioma cell lines and TCGA were used to validate the biological findings. Here, we develop APOLLO (rAman-based PathOLogy of maLignant gliOma)-a computational workflow that predicts different subtypes of glioma from spontaneous Raman spectra of FFPE tissue slides. Our novel APOLLO platform distinguishes tumors from nontumor tissue and identifies novel Raman peaks corresponding to DNA and proteins that are more intense in the tumor. APOLLO differentiates isocitrate dehydrogenase 1 mutant (IDH1mut) from wild-type (IDH1WT) tumors and identifies cholesterol ester levels to be highly abundant in IDHmut glioma. Moreover, APOLLO achieves high discriminative power between finer, clinically relevant glioma methylation subtypes, distinguishing between the CpG island hypermethylated phenotype (G-CIMP)-high and G-CIMP-low molecular phenotypes within the IDH1mut types. Our results demonstrate the potential of label-free Raman spectroscopy to classify glioma subtypes from FFPE slides and to extract meaningful biological information thus opening the door for future applications on these archived tissues in other cancers.

Customizing pyfao56 for evapotranspiration estimation and irrigation scheduling at the Limited Irrigation Research Farm (LIRF), Greeley, Colorado

Estimation of evapotranspiration (ET, the water used by soil evaporation and plant transpiration) and soil water depletion (Dr, the amount of water to bring the soil water in the root zone to field capacity) are critical in irrigation water management. The open-source Python-based crop ET and water balance modeling package named “pyfao56” was originally developed based on the dual crop coefficient approach as described in the Food and Agricultural Organization of the United Nations, Irrigation and Drainage Paper No. 56 (FAO-56). The package later expanded on the seminal FAO-56 document to consider various options relevant to ET and water balance modeling, including both short (grass) and tall (alfalfa) reference crops, optional discretization of variable soil layers based on field capacity, and interpretation of readily available water (RAW) and Dr based on both a dynamic (growing) root zone and maximum root zone. The package requires two input data objects; the “Parameters” class defines variables affecting soil water balance and ET and the “Weather” class specifies relevant meteorological data. Other optional input data objects include the “Irrigation” class for irrigation events, the “Soil_Profile: class for defining stratified soil layer data, and the “Update” class for assimilation of measured data. Additional tools are available for estimating and forecasting standardized reference ET (ETref), providing seasonal water balance summaries, computing goodness-of-fit statistics between measured and modeled values, and visualizing time series plots of daily Dr, ET, and crop coefficients. The current pyfao56 release (v1.2.1) was incorporated into a customized workflow for specific use at the USDA-ARS Limited Irrigation Research Farm (LIRF) in Greeley, Colorado, and named the LIRF Implementation of Pyfao56 (LIRFIP). Field data from 2023 full and limited irrigation field trials were used to demonstrate the functionality of LIRFIP and its customizations and integration of pyfao56 into the LIRF workflow. Specific customizations included use of pyfao56 “customload” functions to input data from various sources, including weather from an on-site micrometerological station (using the application programming interface (API) developed by the maintainers of the Colorado Agricultural Meteorological (CoAgMet) weather station network), irrigation events (via a shared Google docs sheet), field capacity values by plot and soil layer (via an Excel spreadsheet), and measured soil water content as well as basal crop coefficient (Kcb) calculated from fractional canopy cover (fc) (via an SQLite database). To evaluate multiple research plots, LIRFIP was designed to iterate simulations from multiple instances of pyfao56 and to produce customized output summary files and interactive hypertext markup language (html) graphs as an aid for irrigation management decisions for LIRF field trials. This study demonstrated pyfao56 as a useful, flexible, customizable, and repeatable ET-based water balance model by showcasing its integration within a specific computational workflow for irrigation management field research at LIRF. The approach serves as an example for pyfao56 integration in other water management tools as conceived by water managers, researchers, and practitioners worldwide.

ZARP: A user-friendly and versatile RNA-seq analysis workflow

Background RNA sequencing (RNA-seq) is a widely used technique in many scientific studies. Given the plethora of models and software packages that have been developed for processing and analyzing RNA-seq datasets, choosing the most appropriate ones is a time-consuming process that requires an in-depth understanding of the data, as well as of the principles and parameters of each tool. In addition, packages designed for individual tasks are developed in different programming languages and have dependencies of various degrees of complexity, which renders their installation and execution challenging for users with limited computational expertise. Workflow languages and execution engines with support for virtualization and encapsulation options such as containers and Conda environments facilitate these tasks considerably. The resulting computational workflows can then be reliably shared with the scientific community, enhancing reusability and the reproducibility of results as individual analysis steps are becoming more transparent and portable. Methods Here we present ZARP, a general purpose RNA-seq analysis workflow that builds on state-of-the-art software in the field to facilitate the analysis of RNA-seq datasets. ZARP is developed in the Snakemake workflow language and can run locally or in a cluster environment, generating extensive reports not only of the data but also of the options utilized. It is built using modern technologies with the ultimate goal to reduce the hands-on time for bioinformaticians and non-expert users and serve as a template for future workflow development. To this end, we also provide ZARP-cli, a dedicated command-line interface that may make running ZARP on an RNA-seq library of interest as easy as executing a single two-word command. Conclusions ZARP is a powerful RNA-seq analysis workflow that is easy to use even for beginners, built using best software development practices, available under a permissive Open Source license and open to contributions by the scientific community.

Mutational signatures of colorectal cancers according to distinct computational workflows.

Tumor mutational signatures have gained prominence in cancer research, yet the lack of standardized methods hinders reproducibility and robustness. Leveraging colorectal cancer (CRC) as a model, we explored the influence of computational parameters on mutational signature analyses across 230 CRC cell lines and 152 CRC patients. Results were validated in three independent datasets: 483 endometrial cancer patients stratified by mismatch repair (MMR) status, 35 lung cancer patients by smoking status and 12 patient-derived organoids (PDOs) annotated for colibactin exposure. Assessing various bioinformatic tools, reference datasets and input data sizes including whole genome sequencing, whole exome sequencing and a pan-cancer gene panel, we demonstrated significant variability in the results. We report that the use of distinct algorithms and references led to statistically different results, highlighting how arbitrary choices may induce variability in the mutational signature contributions. Furthermore, we found a differential contribution of mutational signatures between coding and intergenic regions and defined the minimum number of somatic variants required for reliable mutational signature assignment. To facilitate the identification of the most suitable workflows, we developed Comparative Mutational Signature analysis on Coding and Extragenic Regions (CoMSCER), a bioinformatic tool which allows researchers to easily perform comparative mutational signature analysis by coupling the results from several tools and public reference datasets and to assess mutational signature contributions in coding and non-coding genomic regions. In conclusion, our study provides a comparative framework to elucidate the impact of distinct computational workflows on mutational signatures.

An MSstats workflow for detecting differentially abundant proteins in large-scale data-independent acquisition mass spectrometry experiments with FragPipe processing.

Technological advances in mass spectrometry and proteomics have made it possible to perform larger-scale and more-complex experiments. The volume and complexity of the resulting data create major challenges for downstream analysis. In particular, next-generation data-independent acquisition (DIA) experiments enable wider proteome coverage than more traditional targeted approaches but require computational workflows that can manage much larger datasets and identify peptide sequences from complex and overlapping spectral features. Data-processing tools such as FragPipe, DIA-NN and Spectronaut have undergone substantial improvements to process spectral features in a reasonable time. Statistical analysis tools are needed to draw meaningful comparisons between experimental samples, but these tools were also originally designed with smaller datasets in mind. This protocol describes an updated version of MSstats that has been adapted to be compatible with large-scale DIA experiments. A very large DIA experiment, processed with FragPipe, is used as an example to demonstrate different MSstats workflows. The choice of workflow depends on the user's computational resources. For datasets that are too large to fit into a standard computer's memory, we demonstrate the use of MSstatsBig, a companion R package to MSstats. The protocol also highlights key decisions that have a major effect on both the results and the processing time of the analysis. The MSstats processing can be expected to take 1-3 h depending on the usage of MSstatsBig. The protocol can be run in the point-and-click graphical user interface MSstatsShiny or implemented with minimal coding expertise in R.

Analysis of 3760 hematologic malignancies reveals rare transcriptomic aberrations of driver genes

BackgroundRare oncogenic driver events, particularly affecting the expression or splicing of driver genes, are suspected to substantially contribute to the large heterogeneity of hematologic malignancies. However, their identification remains challenging.MethodsTo address this issue, we generated the largest dataset to date of matched whole genome sequencing and total RNA sequencing of hematologic malignancies from 3760 patients spanning 24 disease entities. Taking advantage of our dataset size, we focused on discovering rare regulatory aberrations. Therefore, we called expression and splicing outliers using an extension of the workflow DROP (Detection of RNA Outliers Pipeline) and AbSplice, a variant effect predictor that identifies genetic variants causing aberrant splicing. We next trained a machine learning model integrating these results to prioritize new candidate disease-specific driver genes.ResultsWe found a median of seven expression outlier genes, two splicing outlier genes, and two rare splice-affecting variants per sample. Each category showed significant enrichment for already well-characterized driver genes, with odds ratios exceeding three among genes called in more than five samples. On held-out data, our integrative modeling significantly outperformed modeling based solely on genomic data and revealed promising novel candidate driver genes. Remarkably, we found a truncated form of the low density lipoprotein receptor LRP1B transcript to be aberrantly overexpressed in about half of hairy cell leukemia variant (HCL-V) samples and, to a lesser extent, in closely related B-cell neoplasms. This observation, which was confirmed in an independent cohort, suggests LRP1B as a novel marker for a HCL-V subclass and a yet unreported functional role of LRP1B within these rare entities.ConclusionsAltogether, our census of expression and splicing outliers for 24 hematologic malignancy entities and the companion computational workflow constitute unique resources to deepen our understanding of rare oncogenic events in hematologic cancers.

Quantitative profiling N1-methyladenosine (m1A) RNA methylation from Oxford nanopore direct RNA sequencing data

With the recent advanced direct RNA sequencing technique that proposed by the Oxford Nanopore Technologies, RNA modifications can be detected and profiled in a simple and straightforward manner. Majority nanopore-based modification studies were devoted to those popular types such as m6A and pseudouridine. To address current limitations on studying the crucial regulator, m1A modification, we conceived this study. We have developed an integrated computational workflow designed for the detection of m1A modifications from direct RNA sequencing data. This workflow comprises a feature extractor responsible for capturing signal characteristics (such as mean, standard deviations, and length of electric signals), a single molecule-level m1A predictor trained with features extracted from the IVT dataset using classical machine learning algorithms, a confident m1A site selector employing the binomial test to identify statistically significant m1A sites, and an m1A modification rate estimator. Our model achieved accurate molecule-level prediction (Average AUC = 0.9689) and reliable m1A site detection and quantification. To show the feasibility of our workflow, we conducted a study on in vivo transcribed human HEK293 cell line, and the results were carefully annotated and compared with other techniques (i.e., Illumina sequencing-based techniques). We believed that this tool will enabling a comprehensive understanding of the m1A modification and its functional mechanisms within cells and organisms.

AsteRIa enables robust interaction modeling between chromatin modifications and epigenetic readers.

Chromatin, the nucleoprotein complex consisting of DNA and histone proteins, plays a crucial role in regulating gene expression by controlling access to DNA. Chromatin modifications are key players in this regulation, as they help to orchestrate DNA transcription, replication, and repair. These modifications recruit epigenetic 'reader' proteins, which mediate downstream events. Most modifications occur in distinctive combinations within a nucleosome, suggesting that epigenetic information can be encoded in combinatorial chromatin modifications. A detailed understanding of how multiple modifications cooperate in recruiting such proteins has, however, remained largely elusive. Here, we integrate nucleosome affinity purification data with high-throughput quantitative proteomics and hierarchical interaction modeling to estimate combinatorial effects of chromatin modifications on protein recruitment. This is facilitated by the computational workflow asteRIa which combines hierarchical interaction modeling, stability-based model selection, and replicate-consistency checks for astable estimation of Robust Interactions among chromatin modifications. asteRIa identifies several epigenetic reader candidates responding to specific interactions between chromatin modifications. For the polycomb protein CBX8, we independently validate our results using genome-wide ChIP-Seq and bisulphite sequencingdatasets. We provide the first quantitative framework for identifying cooperative effects of chromatin modifications on protein binding.

Computation-guided transcription factor biosensor specificity engineering for adipic acid detection

Transcription factor (TF)-based biosensors that connect small-molecule sensing with readouts such as fluorescence have proven to be useful synthetic biology tools for applications in biotechnology. However, the development of specific TF-based biosensors is hindered by the limited repertoire of TFs specific for molecules of interest since current construction methods rely on a limited set of characterized TFs. In this study, we present an approach for engineering the specificity of TFs through a computation-based workflow using molecular docking that enables targeted alteration of TF ligand specificity. Using this method, we engineer the LysR family BenM TF to alter its specificity from its cognate ligand cis,cis-muconic acid to adipic acid through a single amino acid substitution identified by our computational workflow. When implemented in a cell-free system, the engineered biosensor shows higher ligand sensitivity, expanding the potential applications of this circuit. We further investigate ligand binding through molecular dynamics to analyze the substitution, elucidating the impact of modulating a single amino acid position on the mechanism of BenM ligand binding. This study represents the first application of biomolecular modeling methods for altering BenM specificity and for gaining insights into how mutations influence the structural dynamics of BenM. Such methods can potentially be applied to other TFs to alter specificity and analyze the dynamics responsible for these changes, highlighting the applicability of computational tools for informing experiments. In addition, our developed adipic acid biosensor can be applied for the identification and engineering of enzymes to produce adipic acid.

Asynchronous modeling workflows in CyberWater with on-demand HPC/Cloud access

Workflow management systems (WfMSs) are commonly used to organize/automate sequences of tasks as workflows to accelerate scientific discoveries. During complex workflow modeling, a local interactive workflow environment is desirable, as users usually rely on their rich, local environments for fast prototyping and refinements before they consider using more powerful computing resources. However, existing WfMSs do not simultaneously support local interactive workflow environments and HPC resources. In this paper, we present a mechanism for on-demand access to remote HPC resources from desktop/laptop-based workflow management software to compose, monitor, and analyze scientific workflows in the CyberWater project. CyberWater is an open-data and open-modeling software framework for environmental and water communities. In this work, we extend the open-model, open-data design of CyberWater with on-demand HPC accessing capacity. In particular, we design and implement the LaunchAgent and JobManager, which can be integrated into a local desktop/Laptop environment to allow on-demand usage of remote HPC resources for computational modeling workflows effectively and efficiently. LaunchAgent manages user authentication to remote resources, prepares computation-intensive or data-intensive tasks as batch jobs, submits jobs to remote resources, and monitors quality of services for users. LaunchAgent interacts seamlessly with other components in CyberWater, providing advantages of user-friendly feature-rich desktop software experience and increased computing power through on-demand HPC/Cloud usage. In our evaluations, we demonstrate how a hydrological modeling workflow that consists of both local and remote tasks can be constructed and show that our new on-demand HPC/Cloud usage helps speed up hydrology workflows while allowing asynchronous HPC/Cloud access from workflows using a desktop graphical user interface.

Local Ordering, Distortion, and Redox Activity in (La0.75Sr0.25)(Mn0.25Fe0.25Co0.25Al0.25)O3 Investigated by a Computational Workflow for Compositionally Complex Perovskite Oxides.

Mixing multiple cations can result in a significant configurational entropy, offer a new compositional space with vast tunability, and introduce new computational challenges. For applications such as the two-step solar thermochemical hydrogen (STCH) generation techniques, we demonstrate that using density functional theory (DFT) combined with Metropolis Monte Carlo method (DFT-MC) can efficiently sample the possible cation configurations in compositionally complex perovskite oxide (CCPO) materials, with (La0.75Sr0.25)(Mn0.25Fe0.25Co0.25Al0.25)O3 as an example. In the presence of oxygen vacancies (VO), DFT-MC simulations reveal a significant increase of the local site preference of the cations (short-range ordering), compared to a more random mixing without VO. Co is found to be the redox-active element and the VO is the preferentially generated next to Co due to the stretched Co-O bonds. A clear definition of the vacancy formation energy (Evf) is proposed for CCPO in an ensemble of structures evolved in parallel from independent DFT-MC paths. By combining the distribution of Evf with VO interactions into a statistical model, the oxygen nonstoichiometry (δ), under the STCH thermal reduction and oxidation conditions, is predicted and compared with the experiments. Similar to the experiments, the predicted δ can be used to extract the enthalpy and entropy of reduction using the van't Hoff method, providing direct comparisons with the experimental results. This procedure provides a full predictive workflow for using DFT-MC to obtain possible local ordering or fully random structures, understand the redox activity of each element, and predict the thermodynamic properties of CCPOs, for computational screening and design of these CCPO materials at STCH conditions.

A workflow management system for reproducible and interoperable high-throughput self-driving experiments

To foster self-driving experimentation and address the reproducibility crisis in bioprocess development in a collaborative environment, a modular Workflow Management System (WMS) is required. In this work, a WMS based on Directed Acyclic Graphs that offers a modular and flexible design for plug-and-play integration of computational tools is presented. A case study is used to demonstrate that the implementation of a computational WMS in robotic experimental facilities promotes the application of Findable, Accessible, Interoperable and Re-usable principles, allowing researchers to readily share protocols, models, methods and data. As a proof of concept, we integrated three different computational workflows for online re-design of feeding rates in 24 parallel E. coli fed-batch cultivations producing elastin-like proteins. This approach provides a solid foundation for increasing scientific data generation in robotic experimental facilities, fostering open collaboration, and addressing the challenges of reproducibility in research.

Mibianto: ultra-efficient online microbiome analysis through k-mer based metagenomics.

Quantifying microbiome species and composition from metagenomic assays is often challenging due to its time-consuming nature and computational complexity. In Bioinformatics, k-mer-based approaches were long established to expedite the analysis of large sequencing data and are now widely used to annotate metagenomic data. We make use of k-mer counting techniques for efficient and accurate compositional analysis of microbiota from whole metagenome sequencing. Mibianto solves this problem by operating directly on read files, without manual preprocessing or complete data exchange. It handles diverse sequencing platforms, including short single-end, paired-end, and long read technologies. Our sketch-based workflow significantly reduces the data volume transferred from the user to the server (up to 99.59% size reduction) to subsequently perform taxonomic profiling with enhanced efficiency and privacy. Mibianto offers functionality beyond k-mer quantification; it supports advanced community composition estimation, including diversity, ordination, and differential abundance analysis. Our tool aids in the standardization of computational workflows, thus supporting reproducibility of scientific sequencing studies. It is adaptable to small- and large-scale experimental designs and offers a user-friendly interface, thus making it an invaluable tool for both clinical and research-oriented metagenomic studies. Mibianto is freely available without the need for a login at: https://www.ccb.uni-saarland.de/mibianto.

BASiCS workflow: a step-by-step analysis of expression variability using single cell RNA sequencing data

Cell-to-cell gene expression variability is an inherent feature of complex biological systems, such as immunity and development. Single-cell RNA sequencing is a powerful tool to quantify this heterogeneity, but it is prone to strong technical noise. In this article, we describe a step-by-step computational workflow that uses the BASiCS Bioconductor package to robustly quantify expression variability within and between known groups of cells (such as experimental conditions or cell types). BASiCS uses an integrated framework for data normalisation, technical noise quantification and downstream analyses, propagating statistical uncertainty across these steps. Within a single seemingly homogeneous cell population, BASiCS can identify highly variable genes that exhibit strong heterogeneity as well as lowly variable genes with stable expression. BASiCS also uses a probabilistic decision rule to identify changes in expression variability between cell populations, whilst avoiding confounding effects related to differences in technical noise or in overall abundance. Using a publicly available dataset, we guide users through a complete pipeline that includes preliminary steps for quality control, as well as data exploration using the scater and scran Bioconductor packages. The workflow is accompanied by a Docker image that ensures the reproducibility of our results.

Analysis of the co-rotating vortex pair as a test case for computational aeroacoustics

The co-rotating vortex pair arrangement is a frequently applied test case in computational aeroacoustics. Its flow quantities and far-field radiated sound pressure can be computed analytically, assuming delta-distribution potential vortices. When this configuration is used for verification in a numerical computation, the singular vortex cores cannot be modeled directly; they have to be desingularized, which affects the associated flow and acoustical fields. Closed-form expressions for the aeroacoustic source terms are derived for the Scully vortex model and used for detailed numerical computations of the radiated sound pressure. The effects of desingularization and finite source region size are analyzed in the frequency domain for Lighthill’s equation and the perturbed convective wave equation by means of an efficient numerical approach that computes the convolution of the aeroacoustic source terms and the 2D free-field Green’s function. It is shown that the deviations of the far-field radiated sound pressure from the idealized analytical solution depend on the Mach number, vortex core radius, and truncation limit of the source region. A significant increase in the effective size of the source region due to the desingularization is demonstrated. Minimal error bounds compared to the analytic far-field pressure solution are established. The presented converged numerical results serve as reference solutions for validating implementations of hybrid computational aeroacoustic workflows.

Regional-scale fault-to-structure earthquake simulations with the EQSIM framework: Workflow maturation and computational performance on GPU-accelerated exascale platforms

Continuous advancements in scientific and engineering understanding of earthquake phenomena, combined with the associated development of representative physics-based models, is providing a foundation for high-performance, fault-to-structure earthquake simulations. However, regional-scale applications of high-performance models have been challenged by the computational requirements at the resolutions required for engineering risk assessments. The EarthQuake SIMulation (EQSIM) framework, a software application development under the US Department of Energy (DOE) Exascale Computing Project, is focused on overcoming the existing computational barriers and enabling routine regional-scale simulations at resolutions relevant to a breadth of engineered systems. This multidisciplinary software development—drawing upon expertise in geophysics, engineering, applied math and computer science—is preparing the advanced computational workflow necessary to fully exploit the DOE’s exaflop computer platforms coming online in the 2023 to 2024 timeframe. Achievement of the computational performance required for high-resolution regional models containing upward of hundreds of billions to trillions of model grid points requires numerical efficiency in every phase of a regional simulation. This includes run time start-up and regional model generation, effective distribution of the computational workload across thousands of computer nodes, efficient coupling of regional geophysics and local engineering models, and application-tailored highly efficient transfer, storage, and interrogation of very large volumes of simulation data. This article summarizes the most recent advancements and refinements incorporated in the workflow design for the EQSIM integrated fault-to-structure framework, which are based on extensive numerical testing across multiple graphics processing unit (GPU)-accelerated platforms, and demonstrates the computational performance achieved on the world’s first exaflop computer platform through representative regional-scale earthquake simulations for the San Francisco Bay Area in California, USA.

Abstract PO4-14-09: A computational model of the mechanisms of action of combined endocrine therapy and CDK4/6 inhibition predicts outcome in patients with Luminal B breast cancer

Abstract Background: Hormone receptor-positive/HER2-negative (HR+/HER2-) breast cancer (BC) is clinically and biologically heterogeneous. CDK4/6 inhibitors (CDK4/6i) are proven to be effective in different molecular subtypes of HR+/HER2- BC, including the PAM50 luminal B. However, not all patients benefit to the same extent and new biomarkers for response are needed. The aim of this study was to develop a computational workflow to predict the response of patients with luminal B BC to treatment with CDK4/6i in combination with endocrine therapy (ET). Methods: The main part of the workflow is a computational model representing the mechanisms of action of the drugs and their influence on protein signaling and cell proliferation in the tumor. The model was trained on publicly available data from BC cell lines (Western blot time courses and viability dose-response data). The model was then used to predict response in patients by incorporating gene expression profiles obtained at baseline. The output of the model was a score indicating how strongly the tumor responded to the treatment. Gene expression data determined using the nCounter BC360 panel was available at baseline from two patient cohorts: 1) the CORALLEEN phase II trial which evaluated neoadjuvant ribociclib plus letrozole vs multi-agent chemotherapy in postmenopausal patients with luminal B by PAM50, HR+/HER2- early BC (Prat et al. Lancet Oncology. 2020), 2) a retrospective study of postmenopausal patients with HR+/HER2-/Luminal B advanced BC treated with CDK4/6i plus ET in the first line setting at Hospital Clinic Barcelona (hereafter CDK cohort). PAM50 risk of relapse (ROR) and Ki67 levels were considered as outcome for the CORALLEEN patients. Progression free survival (PFS) was available in the CDK cohort. Area under the ROC Curve (AUC) was used to estimate the discrimination performance of the model. Differences were tested for statistical significance using Wilcoxon rank-sum test. Hazard ratios were estimated using Cox regression to investigate the association with PFS. Results: The computational model used here describes the dynamics of relevant protein-protein and drug-protein interactions to the treatment of CDK4/6i plus ET. The model showed high agreement with the training data obtained from cell lines (Pearson correlation of 0.88 and 0.95). The ability of the trained model to predict treatment outcome in patients was validated in baseline samples of the CDK4/6i arm (n=51) and the chemotherapy arm (n=52) of the CORALLEEN trial, and baseline Luminal B primary tumors of the CDK cohort (n=19). For each patient, a response score was calculated by the model using the expression of 6 genes at baseline as input. This response score was significantly associated with patient response in both cohorts. The model identified patients with high Ki67 (AUC of 0.80 (95% CI 0.64 - 0.92)) and high ROR (AUC of 0.78 (95% CI 0.64 - 0.89)) after treatment in the CDK4/6i arm of the CORALLEEN cohort and was used to stratify patients into different response groups. The AUC in the chemotherapy arm of CORALLEEN for identifying high Ki67 was 0.44 (95% CI 0.29 - 0.58), indicating that the predictions are specific to CDK4/6i plus ET treatment. Additionally, stratification of patients in the CDK cohort into either two or three groups based on the response score was linked to PFS (HR=3.71 (95% CI 0.97 - 14.16), p=0.055 and HR=2.92 (95% CI 1.08 - 7.86), p=0.034, respectively). Conclusion: A mechanistic model was developed, trained on cell line data, and then used to predict treatment outcome of patients with HR+/HER2-/PAM50 Luminal B BC treated with CDK4/6i plus ET. This approach showed significant association with observed outcome and could be used to assign patients to different response groups, indicating the usefulness of the model as a potential marker for response. Citation Format: Leonard Schmiester, Fara Brasó-Maristany, Vessela Kristensen, Arnoldo Frigessi, Aleix Prat, Alvaro Köhn-Luque. A computational model of the mechanisms of action of combined endocrine therapy and CDK4/6 inhibition predicts outcome in patients with Luminal B breast cancer [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO4-14-09.

Abstract PO5-01-09: Exploring cellular heterogeneity of localised breast cancers

Abstract Breast cancer is a heterogeneous disease at multiple levels, ranging from subtype differences between patients (inter-patient heterogeneity) to the diverse composition of malignant cells, heterogeneity of hormone receptor (HR) expression and cellular makeup within single breast cancer samples (intra-tumour heterogeneity). Despite an increase in the number of effective therapies available, many patients will experience an incomplete treatment response and subsequent relapse. These adverse treatment outcomes may be attributed to the often overlooked but critical factor of cellular heterogeneity. In order to gain deeper insights into intra-tumour heterogeneity, we have applied single-cell technologies on a cohort comprising 250 primary, untreated breast cancers. We have optimized methods for tissue cryopreservation, eliminating the need for fresh sample processing, and multiplex tissue profiling. Together, these are cost-efficient processes that allow for improved handling of small tissue sizes, such as biopsies, and reduce batch effect. To ensure accurate and reliable data processing, we have developed a scalable computational workflow that includes benchmarked methods for sample SNP-demultiplexing, doublet detection, high-resolution cell annotation and cellular integration. Finally, we are extending our existing methods1 to study the cellular heterogeneity of breast cancers. Our method, scSubtyper, explores the phenotypic differences between malignant cells within tumours, by comparing each single cell to distinct features associated with different molecular subtypes and assigning each cell to one of these subtypes. Our previous study1 and preliminary results of this project revealed that over 90% of the samples exhibit a mix of malignant cells of different subtypes, and 50% of samples contain cells that have characteristics of all subtypes, demonstrating cellular heterogeneity exists not only exists between malignant cells, but also within malignant cells of a tumour. Our second approach, known as ecotyping, assesses patterns of cell type frequencies across samples and groups them based on similarity of these co-occurences. Our preliminary results have revealed the existence of 5 ecotypes that lack significant associations with samples clinical subtypes. Applying the same approach exclusively within the HR-positive samples identified 4 ecotypes characterized by distinct abundances of immune and stromal cells. This analysis revealed that ecotypes are not a simple surrogate for clinical and molecular subtypes, but their presence could drive different response to treatment. Together, our high-throughput tissue processing and computational approaches to studying intra-tumour heterogeneity are now being applied to our large, well annotated, clinical cohort. Supported by the preliminary results, we hypothesize that this study will play a vital role in optimizing breast cancer patient stratification to improve treatment management and outcome. 1. Wu, Sunny Z., et al. "A single-cell and spatially resolved atlas of human breast cancers." Nature genetics 53.9 (2021): 1334-1347. Citation Format: Beata Kiedik, Daniel Roden, Kate Harvey, Ghamdan Al-Eryani, Sunny Wu, Mun Hui, Sandra O'Toole, Elgene Lim, Charles Perou, Alex Swarbrick. Exploring cellular heterogeneity of localised breast cancers [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO5-01-09.

Machine Learning-Accelerated High-Throughput Computational Screening: Unveiling Bimetallic Nanoparticles with Peroxidase-Like Activity.

Bimetallic nanoparticles (NPs) with peroxidase-like (POD-like) activity play a crucial role in biosensing, disease treatment, environmental management, and other fields. However, their development is impeded by a vast range of tunable properties in components and structures, making the establishment of structure-effect relationships and the discovery of active materials challenging. Addressing this, we established robust scaling relationships by meticulously analyzing the catalytic reaction networks of pure metal NPs, which laid the volcano-shaped correlation between the activity and O* adsorption energy. Utilizing these relationships, we introduced an innovative and versatile descriptor of the NPs, which was then integrated into a machine learning-accelerated high-throughput computational workflow, significantly boosting the predictive accuracy for the POD-like activity of bimetallic NPs. Our methodological approach enabled the successful prediction of activities for 1260 bimetallic NPs, leading to the identification of several highly effective catalysts. Furthermore, we distilled several strategies for designing efficient bimetallic NPs based on our screening results.

SNP-SVant: A Computational Workflow to Predict and Annotate Genomic Variants in Organisms Lacking Benchmarked Variants.

Whole-genome sequencing is widely used to investigate population genomic variation in organisms of interest. Assorted tools have been independently developed to call variants from short-read sequencing data aligned to a reference genome, including single nucleotide polymorphisms (SNPs) and structural variations (SVs). We developed SNP-SVant, an integrated, flexible, and computationally efficient bioinformatic workflow that predicts high-confidence SNPs and SVs in organisms without benchmarked variants, which are traditionally used for distinguishing sequencing errors from real variants. In the absence of these benchmarked datasets, we leverage multiple rounds of statistical recalibration to increase the precision of variant prediction. The SNP-SVant workflow is flexible, with user options to tradeoff accuracy for sensitivity. The workflow predicts SNPs and small insertions and deletions using the Genome Analysis ToolKit (GATK) and predicts SVs using the Genome Rearrangement IDentification Software Suite (GRIDSS), and it culminates in variant annotation using custom scripts. A key utility of SNP-SVant is its scalability. Variant calling is a computationally expensive procedure, and thus, SNP-SVant uses a workflow management system with intermediary checkpoint steps to ensure efficient use of resources by minimizing redundant computations and omitting steps where dependent files are available. SNP-SVant also provides metrics to assess the quality of called variants and converts between VCF and aligned FASTA format outputs to ensure compatibility with downstream tools to calculate selection statistics, which are commonplace in population genomics studies. By accounting for both small and large structural variants, users of this workflow can obtain a wide-ranging view of genomic alterations in an organism of interest. Overall, this workflow advances our capabilities in assessing the functional consequences of different types of genomic alterations, ultimately improving our ability to associate genotypes with phenotypes. © 2024 The Authors. Current Protocols published by Wiley Periodicals LLC. Basic Protocol: Predicting single nucleotide polymorphisms and structural variations Support Protocol 1: Downloading publicly available sequencing data Support Protocol 2: Visualizing variant loci using Integrated Genome Viewer Support Protocol 3: Converting between VCF and aligned FASTA formats.