NCI Data Jamboree (Project Abstract Submission)

Filter submissions
61 submissions
# Created User IP address First Name Middle Initial Last Name Degree(s) Position/Title/Career Status Organization Organization Address Email List of Additional Authors Abstract Category Abstract Keywords Abstract Title Abstract Operations
67 Tue, 07/28/2026 - 12:20 bojae 10.208.24.192 Yuanwei Bay Xu PhD, MSE Postdoc Researcher Johns Hopkins University Baltimore, Maryland yxu105@jh.edu Developing, refining, or validating tools, methods, algorithms, and pipelines Spatial proteomics, CRDC, imaging An Automated Computational Workflow for Co-Registering Spatially-resolved Proteomics with Public Reference Atlases Spatial proteomics technologies generate rich microenvironmental data, yet critical barriers prevent their integration into public multi-omic repositories. This project will develop an open-source computational workflow to harmonize custom grid-based spatial proteomic datasets with standardized reference atlases and multi-modal imaging archives, enabling seamless data sharing within the NCI Cancer Research Data Commons (CRDC) ecosystem.

Specific Aim 1: Implement Image Registration Algorithms for Spatial Coordinate Harmonization. We will develop computational methods to transform SPOTTER-derived spatial proteomic data from horizontal mouse brain sections into standardized coordinate systems, specifically the Allen Mouse Brain Atlas Common Coordinate Framework (CCFv3) and H&E-stained histology references. Using image registration libraries (Valis, SimpleITK), we will generate validated transformation matrices, establish annotation transfer protocols from atlas regions to proteomic measurements, and define quality control metrics for alignment accuracy. This aim addresses the fundamental technical challenge of bridging experimental coordinate spaces with canonical anatomical frameworks.

Specific Aim 2: We will package the registration tools into a cloud-based, containerized workflow that merges spatial proteomics with complementary data modalities and prepares CRDC-compatible submissions. The pipeline will accept custom proteomic datasets, execute automated coordinate alignment, overlay spatial transcriptomics from 10x Visium/HTAN repositories, incorporate whole-slide DICOM pathology from NCI Imaging Data Commons, and output standardized data objects using community frameworks (such as SpatialExperiment, Scanpy, Seurat v5). Comprehensive documentation will enable plug-and-play adoption across platforms.
Expected Outcomes & Impact: This collaborative project will deliver an end-to-end workflow democratizing spatial proteomics data sharing, accelerating multi-omic cancer research through standardized harmonization procedures. All code and example datasets will be released as open source.
63 Mon, 07/27/2026 - 23:11 Anonymous 10.208.24.192 Avani Patel N.A. Animal Genome Institute Palmyra info@animalgenomeinstitute.org
  • First Name: Yuka
    Last Name: Imamura
    Affiliation: Animal Genome Institute
 Employing statistical, computational, and informatics tools, algorithms, and methods to integrate or analyze data Cross-Species Signature Scoring and Risk Assessment Modeling for Human and Canine Osteosarcoma Osteosarcoma is a genetically complex malignancy. While human and canine osteosarcomas share significant molecular similarities, integrating datasets across species and sequencing platforms remains a technical challenge. Overcoming these hurdles is crucial for identifying conserved oncological targets and improving the AI-readiness of comparative genomic data for the broader research community. We aim to develop and validate a robust computational framework for cross-species data integration, specifically focusing on continuous signature scoring models for human and canine osteosarcoma. By utilizing publicly available human osteosarcoma datasets (such as TARGET-OS) and public canine cohorts (such as NCI's DOG² cohort), we will employ expression data scaling techniques, cross-platform normalization, and continuous signature scoring algorithms to harmonize the disparate matrices. Ultimately, we aim to leverage these validated models to develop a cross-species risk assessment application to support clinical decision-making.
62 Mon, 07/27/2026 - 22:56 Anonymous 10.208.24.192 Yi Hsiao Ph.D. Research Fellow University of Michigan Ann Arbor yihsiao@umich.edu
  • First Name: Alexey
    Last Name: Nesvizhskii
    Post-nominal letters: Ph.D.
    Affiliation: University of Michigan
  • First Name: Marcin
    Last Name: Cieslik
    Post-nominal letters: Ph.D.
    Affiliation: University of Michigan
 Employing statistical, computational, and informatics tools, algorithms, and methods to integrate or analyze data Leukemia; AML; multi-omics; biomarker discovery; drug target nomination Building a Harmonized Multi-Omics Resource and Web Application for Biomarker Discovery and Drug Target Nomination in Leukemia Leukemia is highly heterogeneous across genomic, epigenomic, proteomic, metabolic, and clinical dimensions, creating challenges and opportunities for biomarker discovery and therapeutic target nomination. This project aims to develop a harmonized multi-omics resource and web application integrating publicly available leukemia cohorts, using acute myeloid leukemia (AML) as the primary use case.

During the jamboree, we will identify and prioritize relevant leukemia datasets from resources such as CPTAC, TCGA, TARGET, Beat AML and related public repositories. We will initially focus on datasets identified before the data jamboree (more than 10) including CPTAC adult AML datasets and recent CPTAC–Kids First collaboration datasets on pediatric AML and T-cell acute lymphoblastic leukemia. Data modalities of interest include clinical annotations, bulk genomics, transcriptomics, epigenomics (including DNA methylation and ATAC-seq), proteomics, post-translational modifications such as phosphorylation, metabolomics, lipidomics, drug screening, and gene dependencies.

We will use AI-assisted extraction to define unified study- and sample-level annotations and prepare standardized datasets using controlled vocabularies, including OncoTree/NCI Thesaurus for disease classification and HGNC for molecular identifiers. In parallel, we will create a user-friendly web application for exploring cohorts, querying molecular features and candidate targets across omics layers, visualizing feature distributions, and assessing associations with clinical variables such as mutation, subtype, treatment response, relapse, and survival. Because the jamboree phase will focus on data inventory, harmonization, lightweight analysis, and implementing analyses, standard personal computers with internet access, AI-enabled tools, and open-source software will be sufficient.

Key scientific questions include: Which molecular features are associated with leukemia subtype, treatment response, or survival? Can integrated multi-omics profiles nominate actionable biomarkers, dysregulated pathways, or drug targets not apparent from single-omic analyses? What harmonization barriers limit cross-cohort leukemia biomarker and target discovery? The expected outcome is a reusable harmonized resource, documentation, and web application for community-driven leukemia translational discovery.
64 Mon, 07/27/2026 - 21:57 Anonymous 10.208.24.192 Jennifer M Torres Del Valle PhD, MPH, MA Postdoctoral NIH Scholar in Clinical and Translational Research Postdoctoral Master of Science in Clinical and Translational Research Program, University of Puerto Rico Medical Sciences Campus, San Juan, Puerto Rico San Juan, Puerto Rico jennifer.torres@upr.edu
  • First Name: Claudia
    Last Name: Amaya Ardila
    Post-nominal letters: EdD, MPH, MS
    Affiliation: Department of Biostatistics and Epidemiology, Graduate School of Public Health; and Master of Science in Clinical and Translational Research Program, University of Puerto Rico Medical Sciences Campus, San Juan, Puerto Rico
  • First Name: Souhail
    Last Name: Malave Rivera
    Post-nominal letters: PhD, MSc
    Affiliation: Department of Social Sciences, Graduate School of Public Health; and Master of Science in Clinical and Translational Research Program, University of Puerto Rico Medical Sciences Campus, San Juan, Puerto Rico
  • First Name: Maritere
    Last Name: Meléndez Rosario
    Post-nominal letters: DPT, MPH, MS
    Affiliation: Department of Biostatistics and Epidemiology, Graduate School of Public Health, University of Puerto Rico Medical Sciences Campus, San Juan, Puerto Rico
  • First Name: Andrea
    Last Name: AngaritaValderrama
    Post-nominal letters: MPH
    Affiliation: Department of Biostatistics and Epidemiology, Graduate School of Public Health, University of Puerto Rico Medical Sciences Campus, San Juan, Puerto Rico
 Evaluating data quality for reproducibility and AI-readiness HPV-associated cancers, adolescents and young adults, conditional relative survival, cancer survivorship, SEER Research Plus Survival After HPV-Associated Cancer in Adolescents and Young Adults: Changes With Time Since Diagnosis Background: Adolescents and young adults (AYAs) have distinct cancer survivorship needs. Estimates measured from diagnosis do not reveal whether survival differences across human papillomavirus (HPV)-associated cancer sites persist as patients survive additional years. These patterns could inform survivorship research, risk communication, and care planning.

Objective: To determine whether differences in five-year conditional relative survival across HPV-associated cancer sites and stages persist, narrow, or widen after patients survive one, three, and five years.

Approach: The team will conduct a population-based retrospective cohort study using SEER Research Plus Data, 21 Registries. The cohort will include patients aged 15–39 years with a first primary invasive, registry-defined HPV-associated cancer diagnosed during 2000–2013 and followed through December 31, 2023. Established SEER site and histology classifications will define cervical carcinoma, noncervical anogenital squamous cell carcinoma, and oropharyngeal squamous cell carcinoma. Follow-up will extend from diagnosis to death, last contact, or study closure. SEER*Stat will estimate five- and ten-year relative survival and five-year conditional relative survival at diagnosis and among patients surviving one, three, and five years. Absolute differences and relative contrasts will be evaluated across cancer-site groups, harmonized SEER Summary Stage, diagnostic periods, and survival landmarks. Using institutional computing resources, R will support data management and observed-survival analyses with Kaplan–Meier curves, log-rank tests, and multivariable Cox models adjusted for site, stage, age, sex, and diagnostic period. Diagnostics and sensitivity analyses will address proportional hazards, unknown stage, data completeness, and unstable estimates.

Before the Jamboree, the team will finalize cohort specifications, confirm data access and variables, and assess cell sizes and events. During the Jamboree, the team will implement a reproducible workflow, generate preliminary estimates and visualizations, evaluate limitations, and produce reusable cohort specifications and analysis code. A multidisciplinary team with expertise in epidemiology, biostatistics, HPV-related cancer prevention, and clinical and translational research has been assembled.
61 Mon, 07/27/2026 - 21:28 Anonymous 10.208.24.192 Candice Francheska Tambaoan MPH Early Career Researcher N/A Malden, MA cfbt@bu.edu Employing statistical, computational, and informatics tools, algorithms, and methods to integrate or analyze data Real World Data, Genomics, Biomarkers, Disease Monitoring Applying as a Project Seeker Experience and Expertise:
- My background is in molecular biology, genetics, epidemiology, and biostatistics
- I can code in R, SAS, and Python. Can perform longitudinal data analysis, survival analysis, regression models, among others.
- Can wrangle and analyze with real world (e.g. Flatiron, SEER, EHR, TCGA) and clinical trial data.
- Worked as a data scientist for an NGS cancer diagnostic company, mainly focusing on the applications of liquid biopsy in treatment response monitoring. Also worked on the application of comprehensive genomic profiling data in patient selection, resistance monitoring, and biomarker development.
Why I want to participate in the jamboree:
- I would like to join this event because it offers a unique chance to work directly on cancer datasets alongside experienced researchers. It is also a good opportunity to learn from my fellow researchers, especially those coming from different backgrounds. Lastly, I believe that this event can lead to projects that can make substantial impact on patient lives.
What I hope to achieve:
- I hope to join a team where I can contribute my analytical skills and expertise meaningfully.
- I hope to expand my exposure to other data types and repositories
- I hope to build connections within the research community and explore whether the work can lead to a publication or can extend into an on-going collaboration beyond the event itself.
60 Mon, 07/27/2026 - 20:59 Anonymous 10.208.24.192 Trevor R Michelson M.S. Data Scientisti BVARI Boston Trevor.Michelson@va.gov
  • First Name: Alice
    Last Name: Kwak
    Post-nominal letters: PhD
    Affiliation: Department of Veterans Afffairs
  • First Name: Nicholas
    Last Name: Lemmer
    Post-nominal letters: MS
    Affiliation: Department of Veterans Affairs
  • First Name: Nithin
    Last Name: Weerasinghe
    Post-nominal letters: MS
    Affiliation: Department of Veterans Affairs
 Enhancing data interoperability (e.g., data harmonization, data federation) Precision Oncology, Federated Data Integration, VA EHR, Multi-Modal Oncology Data Integrating VA Precision Oncology Data with NCI Public Resources for Large-Scale Federated Cancer Research A crucial challenge facing oncological data analysis is the lack of available data sets that have large enough patient cohorts for gaining insights into specific cancer types. To meet this particular challenge, we propose to leverage the US Department of Veterans Affairs (VA) large-scale electronic patient health record that contains millions of patients spanning over several decades. We plan to develop solutions for integrating VA data with other public data sets available through the NCI, resulting in a large-scale longitudinal multi-modal oncology data set accessible to the larger research community. The Precision Oncology Data Repository (PODR) is a VA data set that aggregates, curates, and shares clinical, imaging, and genomic data from various VA and external partner sources. Specifically, the VA has aggregated a cohort of over 140k decedent oncology patients that includes various cancer types with regulatory approval to de-identify and share with trusted external partners. Our goal is to develop various methods for utilizing PODR with other resources to demonstrate data analysis within a federated infrastructure.

The following NCI public data sets are planned to be used, but not limited to:
- Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial Data
- Lung
- Colorectal
- Questionnaires
- Cancer Epidemiology Descriptive Cohort Database (CEDCD)
- Cancer Prevention Study II Nutrition Cohort (CPS-II Nutrition)
- Millenium Cohort Study Panels 1 – 5
- SEER Research
- Prostate Cancer with Decipher Prostate Genomic Classifier Database
- All of Us Research (AoU) Program

The project will require members with experience in ETL processes, healthcare data and analysis, and using version control systems to help facilitate collaborative development. Potential to explore NLP-derived concept code extraction through clinical notes and surveys.

We will make use of VA internal workspaces (e.g., VINCI, ARCHES) as well as test out the Bridges Cancer Genomics Cloud.
59 Mon, 07/27/2026 - 18:57 Anonymous 10.208.28.116 Rawan Elshobaky B.S. Molecular and Cellular Biology PhD student University of Colorado Anschutz Aurora, Colorado rawan.elshobaky@cuanschutz.edu
  • First Name: .
    Last Name: .
    Affiliation: .
 Employing statistical, computational, and informatics tools, algorithms, and methods to integrate or analyze data multi-omics, risk prediction, precision medicine, computational genomics Integrating Multimodal Data for Improving Risk Prediction I am a Ph.D. student in Human Medical Genomics at the University of Colorado Anschutz. I earned my B.S. in Molecular and Cellular Biology from Johns Hopkins with minors in Computer Science and Bioethics. My doctoral research, under the mentorship of Drs. David Conti and Milton Pividori, focuses on developing computational methods to improve the performance, interpretability, and portability of polygenic risk scores (PRS) across populations. Specifically, I use data from the UK Biobank to develop and evaluate risk scores based on predicted gene expression data and utilize functional information with Bayesian fine-mapping methods to prioritize likely causal variants and interpret genetic associations.
I am excited to participate in the NCI Data Jamboree because its team-based, interdisciplinary format provides a unique opportunity to work alongside researchers with complementary expertise to address challenges in cancer data integration and analysis. My research is motivated by the need to develop computational methods that perform reliably across populations, recognizing that robust genetic prediction depends on developing and evaluating methods using datasets that are representative of the populations they are intended to serve. Therefore, I am particularly interested in projects involving the All of Us Research Program, because its large, broadly representative participant cohort, combined with genomic, multi-omic, environmental, and longitudinal clinical data, provides a unique opportunity to evaluate methods that generalize robustly across populations.
Through the jamboree, I hope to gain practical experience in effectively accessing and leveraging large-scale cancer resources, such as All of Us, and contribute to incorporating multimodal data into reproducible computational workflows. I also hope to bring this expertise back to the University of Colorado Anschutz Department of Biomedical Informatics and help facilitate the adoption of these resources in future collaborative research. Finally, I look forward to building lasting collaborations with other participants and continuing these interactions beyond the jamboree.
58 Mon, 07/27/2026 - 17:28 Anonymous 10.208.28.116 Calvin Tribby PhD, MPH City of Hope Comprehensive Cancer Center Duarte, CA ctribby@coh.org
  • First Name: Marta
    Last Name: Jankowska
    Post-nominal letters: PhD
    Affiliation: City of Hope Comprehensive Cancer Center
  • First Name: Jiue-An
    Last Name: Yang
    Post-nominal letters: PhD
    Affiliation: City of Hope Comprehensive Cancer Center
 Employing statistical, computational, and informatics tools, algorithms, and methods to integrate or analyze data geospatial decision support; interactive visualization; spatial epidemiology; community engagement Cancer Screening Priority Mapper: A Flexible Multi-Scale Geospatial Decision Support Tool for Identifying High-Priority Cancer Screening Areas Cancer centers, public health agencies, and community organizations often use different criteria to identify areas with the greatest need for cancer screening interventions. However, there is currently no standardized, flexible tool that allows users to compare how priority areas change when different screening, cancer burden, healthcare access, and sociodemographic indicators are emphasized. This project will develop a publicly accessible web mapping application that integrates county-level measures of cancer burden with tract-level measures of screening barriers and access to care to visualize high-priority areas for screen-detectable cancers across the United States. The project aligns with the NCI Data Jamboree goal of promoting data harmonization, integration, visualization, and analysis by creating a reproducible framework for combining multiple cancer-related datasets into an interactive decision-support tool. Datasets will include county-level cancer incidence and late-stage incidence measures from State Cancer Profiles (NCI and CDC) for breast, cervical, colorectal, lung, and prostate cancers. Cancer screening estimates will be from CDC PLACES. Measures of access will be derived from the ACR Lung Cancer Screening Registry, National Provider Identifier (NPI/NPPES) data, and FDA-certified mammography facility locations. Additional indicators describing sociodemographic characteristics, transportation access, health insurance coverage, and economic conditions will be from U.S. Census data. The resulting application will allow users to select indicators, assign priorities, and visualize how geographic targeting recommendations change under alternative decision-making frameworks. This functionality will support resource allocation, outreach planning, community engagement, and cancer screening program development. During the Jamboree, the team will develop automated data pipelines using APIs, integrate datasets into a unified geospatial database, design the application's prioritization framework and user interface, and optimize analytical workflows for rapid data processing and visualization. Team members with expertise in APIs, dashboard or mapping applications, R/Python, cancer risk factors, cancer screening, small area estimates, or cancer statistics are welcome to join.
56 Mon, 07/27/2026 - 16:49 Anonymous 10.208.28.116 Nora L. Nock Ph.D., MS, BS, PE, FSBM Professor Case Western Reserve University Cleveland nln@case.edu
  • First Name: Mireya
    Last Name: Diaz-Insua
    Post-nominal letters: PhD
    Affiliation: Case Western Reserve University
  • First Name: Siran
    Last Name: Koroukian
    Post-nominal letters: PhD
    Affiliation: Case Western Reserve University
  • First Name: Johnie
    Last Name: Rose
    Post-nominal letters: MD, PhD
    Affiliation: Case Western Reserve University
 Evaluating data quality for reproducibility and AI-readiness AI, qualitative data, lived experience, interviews, focus groups Leveraging AI to Augment Human Expertise and Enhance Efficiency in Synthesizing and Analyzing Lived Experience Qualitative Data in Cancer Research Lived experience obtained from cancer patients, caregivers, families and providers can help integrate cultural context, optimize intervention design, improve clinical outcomes and better understand attitudes, barriers and facilitators to widespread dissemination and implementation efforts. The increased use of qualitative methods in cancer research has created an imminent need to develop more standardized workflows and methods to comprehensively and efficiently synthesize and analyze qualitative data. Leveraging AI to synthesize lived experience qualitative data may help improve efficiency in the arduous and time consuming process of coding and thematic analyses and, potentially uncover new insights not initially observed. We propose to synthesize existing guidebooks and raw transcript text data from interviews and focus groups in cancer patients obtained from multiple sources such as published manuscript supplemental files, National Cancer Institute (NCI)-Designated Comprehensive Cancer Centers patient “story” webpages, and other open access platforms with relevant transcript and guidebook data. We propose to use widely available AI tools (e.g., ChatGPT, Claude) as well as more sophisticated “deep” AI tools (e.g., BERT/BERT-Like) to synthesize transcript data, interview and focus group guides, create codebooks and conduct thematic analyses. We propose to evaluate performance using several measures including but not limited to accuracy and reliability/agreement in revealing codes, subcategories and themes compared to those derived from human experts, reflecting readability, lexical diversity, and coherence. We may also explore utilizing different command prompts from prompt engineering frameworks (e.g., RISEN: Role, Instructions, Steps, End Goal, and Narrowing) if time permits. We hypothesize that human interaction and expert oversight will be needed to add cultural context, judgement, and ethical responsibility. We have assembled a team with expertise in qualitative and mixed methods in cancer research, biostatistics, general AI and large databases but would be interested in having additional member(s), particularly those with advanced/deep AI tool expertise, join our team.
57 Mon, 07/27/2026 - 16:47 Anonymous 10.208.28.116 Alec Koppel MS, PhD Senior Scientist/Research Professor Johns Hopkins University Applied Physics Labs / Dept. of Applied Math & Stat. Laurel, MD alec.koppel@jhuapl.edu
  • First Name: Mark
    Last Name: Yarchoan
    Post-nominal letters: MD
    Affiliation: Johns Hopkins University School of Medicine
  • First Name: Mari
    Last Name: Nakazawa
    Post-nominal letters: MD
    Affiliation: Johns Hopkins University School of Medicine
 Employing statistical, computational, and informatics tools, algorithms, and methods to integrate or analyze data instrumental variables, longitudinal studies, nonlinear function approximators, feature construction Foundations for Causal Modeling of Immunotherapy Outcomes Immune checkpoint inhibitors can produce durable cancer responses, yet current predictors of benefit remain limited, tumor-centric, and poorly suited to capture the dynamic host–tumor interactions that govern response, resistance, and toxicity. The Johns Hopkins prospective immunotherapy biobank provides a uniquely rich opportunity to address this gap, with longitudinal, multimodal data from more than 400 real-world patients and over 1,100 biospecimens spanning clinical outcomes, ctDNA, CyTOF immune profiling, cytokines, germline genetics, BCR/TCR sequencing, and antibody analyses. However, the complexity, temporal structure, and nonrandom missingness of these data require a new analytic foundation before advanced causal or adaptive modeling can be reliably deployed.

We showcase a new analysis pipeline that will transform this heterogeneous biobank into a confounder-aware, temporally validated modeling resource for immunotherapy outcome prediction. The pipeline will integrate biologically informed feature construction with proximal causal inference concepts to capture latent drivers of treatment response, including tumor burden, immune competence, disease severity, treatment selection effects, and assay ascertainment patterns. Rather than treating missing data as a nuisance, the framework will explicitly model missingness—including missing-not-at-random mechanisms arising from progression, toxicity, dropout, or clinical decision-making—as informative structure relevant to patient trajectories.

The resulting feature maps and missingness-aware representations will be evaluated through multiple temporal train/test splits designed to mimic prospective clinical deployment, with initial emphasis on survival prediction and related immunotherapy outcomes. This approach is novel in combining longitudinal multimodal immune-oncology data integration, proxy-based confounder mitigation, missingness-aware modeling, and temporally robust validation within a single scalable pipeline. By establishing this foundation, the project will enable reliable downstream development of causal recovery, intervention design informed by models of host–tumor dynamics under immune checkpoint blockade. Ultimately, this work will convert a deeply phenotyped real-world biobank into an actionable computational capability for discovering clinically meaningful predictors of immunotherapy benefit and harm.
53 Mon, 07/27/2026 - 15:46 Anonymous 10.208.28.116 Maya Zuhl MSc Lead Software Engineer ICF Rockville Maryland maya.zuhl@icf.com
  • First Name: Ratna
    Last Name: Thangudu
    Post-nominal letters: PhD
    Affiliation: ICF
  • First Name: Alexander
    Last Name: Pilozzi
    Post-nominal letters: MSc
    Affiliation: ICF
  • First Name: Yin
    Last Name: Lu
    Post-nominal letters: PhD
    Affiliation: ICF
 Developing, refining, or validating tools, methods, algorithms, and pipelines Biomedical Multimodal Data Integration, Explainable AI, Agentic AI, Multimodal Reasoning, Translational Cancer Research AI-Assisted Translational Exploration: Connecting Scientific Publications, Multi-omic Data, and Clinical Trials Cancer researchers routinely move between publications, genomic and proteomic datasets, imaging repositories, and clinical trial resources to interpret findings and assess translational relevance. Although these resources are increasingly rich and accessible, they remain largely disconnected, requiring researchers to manually synthesize evidence across multiple portals before identifying clinically actionable insights. This fragmentation limits the effective reuse of publicly available cancer data and slows translation of research findings into potential clinical applications.
This project proposes an AI-assisted translational exploration workflow that seamlessly connects scientific publications with underlying multimodal datasets and relevant clinical trials. Building on our existing BioInsight platform, the workflow extends publication-centered exploration by integrating ClinicalTrials.gov into an explainable, agentic framework. Starting from a publication or biomarker, the system retrieves associated multi-omic evidence, summarizes key biological findings, identifies relevant biomarkers and pathways, and recommends related clinical trials with evidence-based explanations. Rather than serving as a search interface, the workflow demonstrates transparent AI-assisted reasoning across heterogeneous biomedical resources through citations, supporting evidence, and provenance.
The project integrates publicly available scientific publications with cancer datasets from the Cancer Research Data Commons (CRDC), including the Genomic Data Commons (GDC), Proteomic Data Commons (PDC), and Imaging Data Commons (IDC), together with ClinicalTrials.gov. The initial demonstration focuses on publication-centered exploration across genomics, proteomics, imaging, and clinical trials, with an extensible architecture that supports additional modalities.
The prototype leverages hybrid Retrieval-Augmented Generation (RAG), agentic AI workflows, biomedical knowledge integration, and cloud-native computing to enable seamless navigation from scientific evidence to clinically relevant studies within a conversational interface. Emphasizing explainable multimodal reasoning over information retrieval alone, the project aligns with the Data Jamboree goals of advancing data discovery, integration, and reuse across cancer data resources. Building on an existing platform and publicly available data, a functional end-to-end prototype can be completed during the three-day Jamboree.
54 Mon, 07/27/2026 - 15:45 Anonymous 10.208.24.192 Mayanka Chandra Shekar Ph.D. Research Scientist Oak Ridge National Laboratory Knoxville chandrashekm@ornl.gov
  • First Name: Xi
    Last Name: Zhang
    Post-nominal letters: Ph.D.
    Affiliation: Oak Ridge National Laboratory
  • First Name: Ankita
    Last Name: Paul
    Post-nominal letters: Ph.D.
    Affiliation: Oak Ridge National Laboratory
  • First Name: Ethan
    Last Name: Seefried
    Post-nominal letters: Ph.D.
    Affiliation: Oak Ridge National Laboratory
 Employing statistical, computational, and informatics tools, algorithms, and methods to integrate or analyze data whole-slide imaging; foundation-model embeddings; Hierarchical, Spatially Aware Whole-Slide Image Classification for Pediatric Cancer Using RADIANCE Pediatric cancers are rare and morphologically heterogeneous, making it difficult to develop robust whole-slide image classification models. RADIANCE is a pathology data project that creates reusable foundation-model embeddings and links them to patch coordinates, slide and specimen identifiers, model provenance, and available clinical metadata.
During the 3-days, we will evaluate whether embeddings contain sufficient information for pediatric cancer classification. Using whole-slide images from the CCDI MCI, we will select a cohort with suitable diagnostic labels and train a model to classify cancer or histologic subtypes. We will first establish a baseline that treats patch embeddings as an unordered collection. We will then evaluate a hierarchical approach that incorporates patch locations and relationships among neighboring tissue regions. The primary technical question is whether spatially organized aggregation improves classification compared with conventional non-spatial aggregation.
This project is relevant to the broader cancer data community because generating foundation-model embeddings from whole-slide images is computationally expensive. Evaluating reusable embeddings for a concrete downstream task will help determine whether they can support multiple studies without repeatedly processing the source images. We also intend to share the resulting vector database as a resource for the community to use in whole-slide image downstream tasks. The workflow may also provide a template for classification, cohort discovery, case retrieval, and other pathology applications.
The work will require expertise in computational pathology, pediatric cancer, machine learning, and data engineering. Expected tools include Python, PyTorch, pathology foundation-model embeddings, the Milvus vector database for embedding storage and retrieval, and GPU-enabled computing on OLCF’s Frontier supercomputing platform for model training and evaluation. We have assembled a ready-to-go team from Oak Ridge National Laboratory.
Mayanka ChandraShekar (in-person), Ethan Seefried (in-person), Xi Zhang (online), Ankita Paul (online)
47 Mon, 07/27/2026 - 15:30 Anonymous 10.208.28.116 SANDHYA Prabhakaran PhD Research Scientist/ESI Moffitt Cancer center/Integrated Math Onco department Tampa/Florida sandhyaprabhakaran@gmail.com Project seeker computational biology, math modeling, machine learning -- My training is in building computational and mathematical models to better understand 'why' we see the patterns in data. I am interested to learn from others in this cohort on 1) building models to analyze multimodal data, and increase the clinical interpretability; and 2) building the relevant study cohorts to enable (1).
46 Mon, 07/27/2026 - 15:26 Anonymous 10.208.28.116 Katina H. Singh B.S. Research Assistant Memorial Sloan Kettering Cancer Center New York singhk8@mskcc.org Employing statistical, computational, and informatics tools, algorithms, and methods to integrate or analyze data At the Intersection of Immune Monitoring and Statistical Modeling As a Research Assistant for the Bone Marrow Transplant Service at Memorial Sloan Kettering Cancer Center, I perform novel immunoassays and link high throughput biomarker data to clinical outcomes. These patients are aggregated to form cohorts in institution led initiatives and in multi-center trials for later evaluation of immunotherapies and to determine impact of conditioning regimen on success of hematopoietic stem cell transplants. Having the opportunity to work on projects across different oncology services here influenced my pursuit of an MPH in Epidemiology and Biostatistics, where I was one of three Master's students permitted to take PhD-level coursework in causal inference.

My introduction to using observational data analysis to isolate treatment effects in complex care plans this past semester motivated me to investigate ways I could apply these frameworks to our patient cohorts. I plan to bridge my computational and biological expertise through upcoming classes in Bayesian statistics and pathophysiology, allowing me to better isolate the complex factors that modulate cancer treatments. The Jamboree is the ideal venue to meld these methods with my insights gained from working in a premier cancer center, translating institutional experience into solutions for national oncology challenges.

Furthermore, this event offers a chance to directly collaborate with leading experts and garner hands-on analysis experience in handling large, diverse datasets. I am particularly excited to utilize the All of Us dataset to identify ways to make personalized healthcare accessible to everyone and examine the role of environmental factors on health for specific communities. Beyond the technical work, I look forward to building lasting connections with other researchers and gaining exclusive mentoring opportunities. It would be amazing to create something what would facilitate greater discoveries in the field and to in some small way, meaningfully contribute to the public effort in the fight against cancer.
55 Mon, 07/27/2026 - 15:24 Anonymous 10.208.28.116 Giuseppe Tarantino Ph.D. Instructor in Medicine Dana-Farber Cancer Institute Boston giuseppe_tarantino@dfci.harvard.edu
  • First Name: Tyler
    Last Name: Aprati
    Affiliation: Dana-Farber Cancer Institute
  • First Name: Bojan
    Last Name: Karlas
    Affiliation: Dana-Farber Cancer Institute
  • First Name: Paulina
    Last Name: Koehler
    Affiliation: Dana-Farber Cancer Institute
  • First Name: Hyeon-Tae
    Last Name: Hwang
    Affiliation: Dana-Farber Cancer Institute
  • First Name: Cathering
    Last Name: Feng
    Affiliation: Dana-Farber Cancer Institute
  • First Name: Xuelu
    Last Name: Liu
    Affiliation: Dana-Farber Cancer Institute
 Developing, refining, or validating tools, methods, algorithms, and pipelines computational pathology; whole-slide imaging; unsupervised tile clustering; tumor cell states; A pipeline and interactive tool for identifying histomorphological tile clusters associated with tumor transcriptional states Motivation: Transcriptomics has uncovered different biological tumor states that are often associated with disease outcomes and response to therapy; however, transcriptomic sequencing is rarely available in clinical practice.
On the other hand, whole-slide images (WSIs) such as H&E are routinely performed, but are not currently used to derive tumor state information. Linking morphological patterns from WSIs with existing transcriptional programs would allow these tumor states to be identified in routine clinical practice, integrating transcriptomic findings into the precision medicine pipeline.
We aim to develop a general, end-to-end pipeline and an interactive application that quantifies and tests associations between morphological features and features including any transcriptional signature or patient survival.

Approach: For WSIs, the pipeline will embed and group images into morphological clusters to summarize each slide as a vector. For transcriptomic samples, gene expression signature scores are computed directly. We will then develop a predictive model (adjusted for clinical and genomic confounders) associating these morphological features and signature scores from the matched H&E and bulk RNA-seq samples available in TCGA.
Downstream, cluster presence will be linked to outcomes such as overall survival, therapy response. The approach is cohort- and tumor-type agnostic, providing a pipeline from histomorphology of clinical samples to associations with tumor states and survival for hypothesis generation or publication.

Implementation: The goal is to deliver this application as a simple web app. Users paste a gene list or upload a signature file. The tool computes expression scores and returns per cluster results (survival statistics, visual summaries, etc). Interchangeable configurations (slide type, cluster number, alignment, threshold) are supported, and a precomputed library of signatures is browsable alongside user-defined ones.
49 Mon, 07/27/2026 - 15:22 Anonymous 10.208.28.116 Nikita Thakur Master of Science, Computer Science Research Data Analyst (Analytics and AI lead) University of Illinois Cancer Center Chicago nthaku3@uic.edu
  • First Name: Saransh
    Last Name: Singh
    Affiliation: University of Illinois Cancer Center
  • First Name: Juhi
    Last Name: Anand
    Affiliation: University of Illinois Cancer Center
  • First Name: Lakshmi Sravya
    Last Name: Rachakonda
    Affiliation: University of Illinois Cancer Center
 Evaluating data quality for reproducibility and AI-readiness AI-readiness; data quality; Cancer Research Data Commons; multimodal oncology data; data leakage AI-Readiness Scorecard: A Task-Relative Profiler for Multimodal Cancer Research Data Commons Datasets Rationale: NIH Bridge2AI criteria establish that FAIR conformance alone does not make a dataset AI-ready, yet the criterion most likely to invalidate a downstream model, Characterization, remains largely unautomated. Researchers assembling oncology AI cohorts evaluate batch effects, informative missingness, leakage risk, and clinical-variable completeness ad hoc and per repository, because CRDC Commons were developed independently with distinct data models.
Objective: Prototype an installable Python profiler answering one question: Does this oncology dataset support this AI task, and if not, what is missing?
Approach: Repository adapters reduce heterogeneous inputs to a shared representation of feature matrix, sample metadata, and resolved site label. Two check families run against it. Modality-agnostic integrity checks cover missingness and its correlation with site, class imbalance, batch, and site effects via cross-validated site-predictability under a permutation null, and leakage from patient-straddle or identifier columns. Oncology completeness checks cover staging, histology, treatment, outcomes, actionable biomarkers, and ICD-O/mCODE conformance. Domain and task knowledge live in declarative YAML packs rather than code, keeping the engine small and extensible. Output is an evidence-linked data card plus Croissant metadata.
Datasets: GDC TCGA-BRCA (clinical, expression, mutations) and one IDC imaging collection; PDC CPTAC proteomics and SEER as stretch targets. Open access tiers throughout.
Jamboree deliverables: A public repository containing the intermediate representation and adapter interface, working batch-effect and leakage checks validated on TCGA-BRCA, a first oncology completeness pack, and a per-patient multimodal linkage matrix spanning GDC and IDC, with a rendered data card.
Tools and environment: Python (pandas, scikit-learn, pydicom), open-tier repository APIs, and modest compute: the design is metadata-first and requires no pixel-level processing.
Team: Assembled and ready (see additional authors), four members spanning computer science and healthcare data analytics, with roles mapped to architecture layers: representation, integrity checks, oncology pack, and adapters.
48 Mon, 07/27/2026 - 15:11 Anonymous 10.208.28.116 Anthony Cristillo Ph.D., MBA Senior Vice President and Federal Health Lead Revolutional, LLC McLean anthony.cristillo@revolutional.com
  • First Name: Diana
    Last Name: Castiblanco
    Post-nominal letters: Ph.D., FMACP, FNC
    Affiliation: Revolutional, LLC
  • First Name: Rod
    Last Name: Fontecilla
    Post-nominal letters: Ph.D.
    Affiliation: Revolutional, LLC
  • First Name: Naga
    Last Name: Nandivelugu
    Post-nominal letters: B.E., MBA
    Affiliation: Revolutional, LLC
Enhancing data interoperability (e.g., data harmonization, data federation) Data interoperability, Data harmonization, AI readiness, Business rule extraction, Semantic data model AI-Assisted Business Rule Extraction To Improve Interoperability across Oncology Datasets, and Enable Accurate, Reproducible, and Trustworthy Analyses Background: Interpretation of oncology treatment, response, and safety data, including line of therapy, adverse events, treatment response, disease progression, censoring, and evidence status, depends on business rules embedded within study protocols, data dictionaries, case report forms, and statistical analysis plans rather than explicit, computable data elements. These implicit rules limit interoperability across datasets and force AI systems to infer clinical meaning that was never formally represented, reducing reproducibility and trustworthiness. The value of explicit rule representation has precedent in oncology; for example, iRECIST formalized immune-related response criteria beyond RECIST, improving consistency in clinical trial interpretation.

Objective: Evaluate whether AI-assisted extraction and explicit representation of oncology business rules improve cross-source interpretability, mapping accuracy, and reproducibility of downstream analyses.

Methods: Using datasets from the NCI Clinical and Translational Data Commons (CTDC), we will leverage our FedRAMP/NIST800-53 AI-powered provisional patented platform (RISE) to identify business rules from heterogeneous clinical and research artifacts prior to data ingestion and integration. Extracted rules will be mapped to a lightweight canonical model consisting of nine core entities (e.g., Patient, Diagnosis, Treatment Episode, Exposure, Adverse Event, Response Assessment, Disease Progression, Outcome, and Evidence Source) and nine categories of governing business rules. Each rule will retain provenance, source version, extraction confidence, and expert review status. Cancer domain experts will validate the extracted rules to ensure factual grounding and clinical accuracy.

Expected Outcomes: This project does not seek to replace existing oncology standards or create a universal cancer data model. Instead, it evaluates whether explicit, computable representation of business rules provides a semantic layer that complements existing standards, improves interoperability across oncology datasets, and enables more accurate, reproducible, and trustworthy AI-enabled analyses. These findings will inform scalable approaches for semantic harmonization within the cancer research ecosystem.
45 Mon, 07/27/2026 - 13:01 Anonymous 10.208.28.116 Hairong Wang Ph.D. Assistant Professor the University of Texas at Austin Austin hairong@utexas.edu Developing, refining, or validating tools, methods, algorithms, and pipelines Project Seeker Statement: Medical Imaging AI, Multimodal Learning, and Longitudinal Cancer Modeling I am an Assistant Professor in Operations Research and Industrial Engineering at The University of Texas at Austin. My research focuses on artificial intelligence and machine learning for cancer research, particularly medical imaging, multimodal data integration, longitudinal disease modeling, and clinically informed prediction. I have worked on projects involving glioblastoma, liver cancer, pediatric brain tumors, and radiogenomic modeling. My expertise includes deep learning, generative modeling, uncertainty quantification, model evaluation, and the design of clinically meaningful AI studies.

I am interested in joining projects involving medical imaging, longitudinal or multimodal cancer data, data integration, and the evaluation of data-driven methods using publicly available cancer datasets. I can contribute to scientific question formulation, study design, model and evaluation strategy, interpretation of imaging and clinical endpoints, and hands-on computational analysis as needed.

I hope to learn more about NCI-supported data resources and data-sharing infrastructure, collaborate with researchers from complementary backgrounds, and contribute to a focused project that can produce a reproducible analysis, prototype, or framework during the jamboree. I would also be interested in continuing productive collaborations after the event when appropriate.
44 Mon, 07/27/2026 - 12:12 Anonymous 10.208.24.192 Johanna Goderre MPH Health Data Scientist National Cancer Institute Rockville johanna.goderrejones@nih.gov
  • First Name: Radu
    Last Name: Robotin
    Post-nominal letters: PhD
    Affiliation: NCI
  • First Name: Haibin
    Last Name: Wu
    Post-nominal letters: PhD
    Affiliation: NCI
  • First Name: Anne-Michelle
    Last Name: Noone
    Post-nominal letters: PhD
    Affiliation: NCI
  • First Name: Huann-Sheng
    Last Name: Chen
    Post-nominal letters: PhD
    Affiliation: NCI
 Enhancing data interoperability (e.g., data harmonization, data federation) Synthetic data; Cancer registry; Pediatric cancer; Privacy-preserving data; Artificial intelligence Developing a Metadata-Driven Framework for Synthetic NCCR Data to Enable Open Science and AI Development The National Childhood Cancer Registry (NCCR) Data Platform links de-identified, participant-level data across population-based cancer registries, Children's Oncology Group (COG) studies, medical and pharmacy claims, radiation oncology, and area-based measures for children, adolescents, and young adults with cancer. These linked data are an invaluable research resource, yet privacy and governance requirements necessarily limit access for students, educators, software developers, and artificial intelligence (AI) tools. This project proposes a metadata-driven framework for generating open-access synthetic NCCR data that preserves the statistical characteristics of the underlying data while containing no real patient records.
We build on the platform's published, machine-readable metadata: the permissible values and frequencies for every variable. We have already developed a working prototype that generates synthetic registry cohorts directly from this public metadata and enforces clinical consistency rules, producing a fidelity scorecard and a privacy-by-design guarantee. Data will be generated only with aggregate frequencies, so no individual can be re-identified. During the three-day jamboree, the team will extend this prototype and, using authorized aggregate analyses of real NCCR data, calibrate the cross-variable relationships (demographics, disease distributions, treatment patterns, and survival) that metadata alone cannot capture.
Deliverables include the working generation pipeline, a mapping of metadata elements to synthesis methods, an initial validation framework spanning statistical fidelity, clinical plausibility, and disclosure risk, and recommendations for producing AI-ready synthetic datasets suitable for open release. The broader goal is a roadmap for representative synthetic NCCR data that supports education, workforce development, software testing, reproducible research, and responsible AI, without exposing protected information. By calibrating and evaluating against the real platform, the resulting datasets will better reflect the complexity of childhood cancer care while remaining safe to share, enabling researchers, developers, and AI tools to build and benchmark methods without access to sensitive participant-level data.
43 Mon, 07/27/2026 - 11:32 Anonymous 10.208.28.116 Ying Ding Ph.D. University of Texas at Austin Austin ying.ding@austin.utexas.edu
  • First Name: Hairong
    Last Name: Wang
    Affiliation: University of Texas at Austin
 Developing, refining, or validating tools, methods, algorithms, and pipelines Concept-Guided Multimodal Mixture of Experts for Interpretable Computational Pathology Healthcare models are transitioning from unimodal prediction toward multimodal reasoning over heterogeneous diagnostic inputs. In computational pathology, for complex tumor subtypes where morphology alone can be challenging to distinguish, pathology reports and molecular measurements may provide additional diagnostic
5 evidence alongside whole-slide images, yet existing models often fail to clarify how diverse signals assemble into recognizable diagnostic concepts. We propose ConceptM3oE (Concept Multimodal MoE), which embeds concept formation directly within interaction-aware mixture-of-experts (MoE) pathways. The architecture decomposes evidence into modality-specific, redundant, and synergistic experts, which are then projected into structured concept bottlenecks mapping convergence consistent with the regularizing effect of concept learning. This work offers a scalable path toward high-performance medical AI that is inherently verifiable and better aligned with the complex decision-making of clinical practice.