Knowledge in Data Science Symposium (KIDS25)

• Quantitative Live-Cell Imaging for Comprehensive Analysis of Phenotypic Heterogeneity
  Anand Singh, PhD and Zeliha Kilic, PhD  
• Bridging the gap in histopathology image analysis of pediatric solid tumors with unsupervised representation learning
  Ali Foroughi Pour, PhD
• Mapping Nuclear Remodeling in Neuronal Differentiation with Multimodal Imaging and Quantitative Analysis
  Kirby Campbell, Ph.D.
• An AI-Guided Platform for Scalable Expert Annotation of Pediatric Cancer Pathology Images
  Haoran Chen, Ph.D.

Oral Session 1B: Mitochondrial Single Cell Omics
Moderator: Julie Waits, PhD

• Mitovolve Models the Evolution of the Prevalence of Mitochondrial Mutations in Pediatric Leukemia
  Stanley Pounds, PhD
• Deciphering the role of mitochondrial DNA somatic mutations in leukemia tumor evolution
  Gang Wu, PhD
• NetBID identifies hidden drivers and pathways associated with pathogenic mitochondrial DNA variants in leukemia
  Qingfei Pan, PhD
• Custom Genomic Pipelines Reveal New Insights into Leukemia Mitochondrial Genetics
  Mondira Kundu, MD, PhD

Oral Session 1C: Advancing Predictive Medicine Through Longitudinal Clinical Data Analysis
Moderator: Sadie Mirzaei, PhD

• Machine Learning Using High-Dimensional Longitudinal Clinical Data Improves the Prediction of Acute and Chronic Graft-versus-Host Disease after Allogeneic Hematopoietic Cell Transplantation
  Dinesh Keerthi 
• Early-Life Prediction of Hospitalization and Emergency Department Visits in Patients with Sickle Cell Disease Using Laboratory and Social Determinants of Health (SDoH) Data
  Chengzhou Wu, PhD
• Temporal Deep Learning for Longitudinal Modeling of Fetal Hemoglobin in Sickle Cell Disease
  Minzhang Zheng
• Turning Data into Action: Real-Time Patient Tracking to Prevent Treatment Abandonment in Pediatric Oncology in El Salvador
  Fuentes-Alabi de Aparicio, MD, MPH

Oral Session 2A: Building Reliable and Interpretable AI Systems for Medical Image Analysis
Renato Umeton, PhD, Moderator

• Often Wrong, Never in Doubt: GPT-4oV’s Performance Vs. Confidence and Explanations for Head CT Hemorrhage Interpretation
  Briana Malik, BS
• Region-Aware Reconstruction Strategy for Pretraining fMRI Foundation Model
  Ruthwik Reddy Doodipala, B.Tech, MS Candidate
• Bayesian Joint Mediation Model to Identify Brain Regions Mediating Negative Effects of Radiation Therapy on Neurocognition in Medulloblastoma Patients
  Saikat Nandy, PhD
• AI for Automated Body Composition Analysis on Pediatric and Adult CT Images
  Rafah Mresh, MD
• Resiliency to Suboptimal Training Data in Machine Learning Segmentation Models for Radiology
  James Pierce, BS

Oral Session 2B: Single-Cell and Systems Immunology in Pediatric Cancer and Immunotherapy
Jeremy Chase Crawford, PhD, Moderator

• Single-cell immune profiling reveals long-lasting B cell convergence to distinct broadly reactive epitopes following vaccination with chimeric influenza virus hemagglutinins
  Lei Li, PhD
• Cytotoxic bystander activation of endogenous CAR-negative T cells in response to CD19-CAR T cell therapy
  Jia-Hua Qu
• Evaluating CAR T cell response in pediatric brain tumor patients using CSF samples and multi-omics approaches
  Jorge Andres Ibanez Vega, PhD
• An AI agent applied to a pan-pediatric cancer single cell atlas discovers convergent and tumor specific gene expression programs
  Alessandro Davini, MSc
• Single cell transcriptional analysis reveals leukocyte heterogeneity and impacts of JAK1/2 modulation of cytokine signaling in a mouse model of primary hemophagocytic lymphohistiocytosis (pHLH)
  Aidin Foroutan, PhD

Oral Session 2C: From Genomic Insights to Policy Impact: Advancing Pediatric Precision Medicine
Malvika Sharan, PhD, Moderator

• Applying Data Science to Transform National Cancer Policies for Children:  AI x SJCARES Policies
  Desiree Becerra, MPP, MSc
• Large-scale meQTL mapping reveals treatment-related differences in genetic regulation of DNA methylation in survivors of childhood cancer from the St. Jude Lifetime Cohort (SJLIFE)
  Tiffany Eulalio, PhD
• Novel Trajectory-Informed Mendelian Randomization Demonstrates Consistent Causal Protective Effect of Fetal Hemoglobin on Clinical Outcomes in Children with Sickle Cell Disease
  Guolian Kang,PhD

Oral Session 3A: From Data to Drugs: Accelerating Drug Discovery Through Data-Driven Methods
Anang Shelat, PhD and Christoph Gorgulla, PhD, Moderators

• Using Bespoke Data Analysis to Improve Throughput, Communication, and Accuracy in Basic Lab Science: Don’t Forget the Small Things
  Mary Ashley Rimmer, PhD
• Harnessing Data Science to Enable N-of-1 Antisense Therapies in Personalized Medicine
  Davneet Kaur, PhD
• Direct-to-Biology Enabled Molecular Glue Discovery
  Uma Neelakantan
• Developmental dynamics of GPCRs provide insights for drug development for fetal, pediatric, and geriatric population
  Katarina Nemec, PhD, PharmD

Oral Session 3B: AI/ML Modeling in Computational Systems Biology
Stanley Pounds, PhD, Moderator

• Deep Learning Meets High-Throughput CRISPR Editing: Utilizing Novel CHANCE-net and CHANCE-seq platforms for Safe Lead Target Identification
  Jacqueline Chyr. PhD
• AI-driven mRNA optimization for cell-type-specific protein expression
  Steve Mathew D A, Mtech
• Proteome-wide computational analyses reveal links between protein condensate formation and RNA biology
  Swarnendu Tripathi, PhD
• From Literature to Structured Data: Hybrid Manual–AI Curation of the Human Kinome
  Katalin Parej, PhD