Karla Paniagua

My research focuses on the application of deep learning techniques to cancer prognosis through the integration of multi-omics and multi-region sequencing data. I am particularly interested in developing deep learning models that explicitly account for intratumor heterogeneity, a defining feature of cancer progression characterized by the coexistence of genetically distinct subclones within the same tumor. These subclones are major drivers of disease recurrence, metastasis, and treatment resistance, yet they are often overlooked in conventional prognostic models.

My current research centers on non-small cell lung cancer (NSCLC), where improving prognostic accuracy remains a critical clinical need. Despite advances in early detection and treatment, outcomes for many patients remain poor due to the inability of existing tools to distinguish biologically aggressive tumors. By integrating spatially resolved molecular data into advanced deep learning frameworks, my work aims to improve the precision of risk stratification and survival prediction, ultimately enabling more personalized treatment decisions. This approach seeks to enhance existing prognostic tools by reducing both over- and undertreatment, thereby improving patient outcomes.

With the potential to extend to breast cancer and other tumor types, my research contributes to the broader goal of developing clinically applicable, data-driven prognostic models that can be translated into real-world oncology practice to guide personalized therapeutic strategies.

• Yijun Sun, PhD

• PhD, University of Texas at San Antonio, San Antonio, Texas