Traditionally, signaling networks have been studied and dissected in the context of single-cell and single-cell populations. On the other hand, it is also well-established that the cellular microenvironment, which is highly dynamic and heterogeneous, exerts critical influences on signaling networks. However, the community has lacked experimental approaches to study cell-cell interactions and single-cell responses in complex tissues. To address these challenges, the Gujral lab is developing new strategies that will enable studies of signaling networks in the tissue microenvironment.
Organotypic tissue slices as a state-of-the-art-model for mechanistic and drug discovery studies. We have been developing methods for maintaining thin sections of mouse and patient-derived tumor slices for mechanistic and drug discovery studies. These preparations are called organotypic tissue slices, and they preserve the organization and heterogeneity of the cells and extracellular structures within the tumor tissue. The tumor tissue slices are 200 – 400 μm thick, representing ~10-20 layers of cells, and include cancer cells, normal cells, and immune cells in this tumor microenvironment. We have optimized the conditions to maintain the viability of organotypic tumor tissue slices for several weeks in culture. We have also developed methods for the delivery of small molecules using an active flow-based perfusion system. We demonstrated the utility of this ex vivo model system for medium-throughput cytotoxic and immuno-oncology drug screening studies (published in OncoImmunology, 2019, JoVE, 2020, and Lab Chip, 2021, Gut, 2022, Cell Reports Medicine, 2025). In collaboration with Dr. Albert Folch (University of Washington, Bioengineering, our lab is developing approaches to integrate organotypic slices with microfluidics to study tumor-host cell interactions within the organized tumor microenvironment. Read More
Developing New Tools for Phenotypic-Based Drug Discovery
To complement the microtumor-based model system and enhance drug discovery through the integration of signaling networks, our lab has developed tools for systems pharmacology. These tools include multiple machine learning-based platforms (KiR, KInhibition, KiDNN, KiRNet, KinCyte, and KiRHub (under review)) and are used to analyze large-scale drug-target profiling data and predict responses to FDA-approved kinase inhibitors. These tools rank hundreds of drug combinations, pinpoint key kinases, and identify optimal therapeutic strategies for specific cellular phenotypes. By integrating this computational framework with microtumor-based models, we enable personalized treatment strategies, particularly for cancers with limited genomic insights. We have applied these methods to study hard-to-treat and rare cancers (PNAS, 2021; Cell Reports Medicine, 2025; Gut, 2025), host-pathogen interactions (Nature Communications, 2017; PLOS Pathogens, 2023), and immunology (Molecular Systems Biology, 2021; EMBO Mol Med, 2022; eLife, 2023). These studies have uncovered new molecular regulators and therapeutic opportunities, highlighting the potential of our tools for unbiased biological discovery. Read More
Advancing Precision Oncology in Rare Cancers
Rare cancers, which represent over 20% of global cancer diagnoses, are challenging to treat often because a lack of molecular understanding limits therapeutic options. Applying our systems pharmacology-based drug screening platform, along with computational methods and newly developed patient-derived models, has led to the identification of new molecular regulators and potential treatment options for rare cancers, particularly for fibrolamellar carcinoma (FLC) and ependymomas. For FLC, a rare childhood liver cancer, we developed novel preclinical models using patient tissue and microtumor models to create patient-derived xenograft (PDX) models (Cancer Discovery, 2025; Cell Reports Medicine, 2024). Using our systems pharmacology-based screening platform, we identified PLK1 as a critical kinase for FLC tumor growth. This finding was validated in preclinical models in which inhibition of PLK1 significantly reduced FLC tumor growth (Gut, 2025), highlighting PLK1 as a potential therapeutic target. In ependymomas, particularly those with ZFTA-RELAFUS fusions, we identified MERTK as essential for cell viability (preprint). Targeting MERTK signaling may offer new therapeutic strategies for this rare brain tumor. Together, these studies demonstrate how integrating functional screening with patient-derived models can uncover actionable vulnerabilities in rare cancers, paving the way for new therapeutic strategies where few currently exist.
Advancing Preclinical Models and Data Resources for Ultra-Rare and Rare Liver Cancers. A major focus of the lab is continuing efforts to develop preclinical models for ultra-rare and rare cancers, with particular emphasis on rare liver tumors, where progress has been hindered by limited tissue availability and the absence of established models.Through collaborations with rare cancer advocacy groups and cancer centers, we are building a “living” biobank of patient-derived specimens. From these samples, we are creating organoids, patient-derived xenograft (PDX) models, and organotypic tumor slices. When combined with multi-omics analyses, these models will enable the identification of therapeutic targets and deepen our understanding of rare liver tumors. Importantly, we plan to make these resources publicly available to accelerate collaborative research and the development of effective treatments. In addition, these models will be integrated into our systems-based drug repurposing efforts described above. Read More.