Aditya Lahiri

Background: Pediatric brain tumors are the leading cause of disease-related mortality in children, yet many aggressive tumors lack effective therapies. RNA splicing is a hallmark of cancer, but it has not yet been systematically studied in pediatric brain tumors. Methods: We analyzed 729 pediatric brain tumors spanning histologies and molecular subtypes to quantify differential tumor splicing. We developed the Splicing Burden Index (SBI) to enable cross-sample comparisons and performed hierarchical clustering of highly variable splice events to define splicing-informed tumor groups. These were integrated with clinical outcomes, pathway activity, and proteogenomic data. Recurrent splice events were prioritized for predicted functional impact, and in vitro perturbation studies were performed targeting the splicing kinase CDC-like kinase 1 (CLK1). Results: SBI revealed substantial interhistology and intrahistology heterogeneity. Clusters were enriched for histologies and molecular subtypes, several of which were independently associated with survival beyond histology and clinical covariates. Spliceosome pathway activity varied across clusters and was associated with worse survival, yet was not correlated with SBI, indicating distinct dimensions of splicing dysregulation. Functional prioritization identified a recurrent in CLK1 exon 4, required for canonical kinase activity. CLK1 exon 4 inclusion followed an oncofetal pattern and showed context-dependent associations with outcome distinct from total CLK1 expression. Pharmacologic inhibition and exon 4-specific perturbation of CLK1 reduced tumor cell viability and disrupted cancer-relevant splicing and transcriptional programs. Conclusions: This study systematically characterizes splicing in pediatric brain tumors, identifies splicing-informed subgroups, and prioritizes CLK1 exon 4 as an oncofetal tumor-specific event, motivating further preclinical exploration.

Non-Small Cell Lung Cancer (NSCLC) remains a leading cause of cancer-related deaths, often characterized by complex mutational landscapes and resistance to monotherapies. To address this, we developed a novel Boolean network (BN) model that integrates oncogenic and immunotherapy signaling pathways to evaluate the efficacy of FDA-approved drug combinations in NSCLC. This computational framework simulates the behavior of key cancer-related pathways under single and multiple mutation scenarios, offering a system-level understanding of drug response. Our model incorporates sixteen targeted inhibitors and simulates their effects on proliferation-promoting and apoptosis-regulating genes. Drug efficacy was quantitatively assessed using a normalized mean size difference (NMSD) metric. Unlike prior models that examine targeted therapy or immunotherapy in isolation, our integrated approach enables systematic evaluation of synergistic effects between these modalities. Key results show that the inclusion of immunotherapy—particularly PD-L1 inhibitors such as Durvalumab—significantly improves therapeutic outcomes, especially in networks with multiple co-occurring mutations. The most effective four-drug combination identified (Durvalumab + Lumakras + Ribociclib + Capivasertib) targets immune evasion, KRAS signaling, cell cycle regulation, and AKT activation, reducing tumor-promoting signals by 89.2% compared to the untreated state. This study provides a theoretical and mechanistic basis for combining immune checkpoint blockade with targeted therapies in NSCLC and demonstrates the utility of BN modeling in optimizing personalized, mutation-specific treatment strategies.

Background: In 2019, the Open Pediatric Brain Tumor Atlas (OpenPBTA) was created as a global, collaborative open-science initiative to genomically characterize 1,074 pediatric brain tumors and 22 patient-derived cell lines. Here, we present an extension of the OpenPBTA called the Open Pediatric Cancer (OpenPedCan) Project, a harmonized open-source multiomic dataset from 6,112 pediatric cancer patients with 7,096 tumor events across more than 100 histologies. Combined with RNA sequencing (RNA-seq) from the Genotype-Tissue Expression and The Cancer Genome Atlas projects, OpenPedCan contains nearly 48,000 total biospecimens (24,002 tumor and 23,893 normal specimens). Findings: We utilized Gabriella Miller Kids First workflows to harmonize whole-genome sequencing (WGS), whole exome sequencing (WXS), RNA-seq, and Targeted Sequencing datasets to include somatic SNVs, indels, copy number variants, structural variants, RNA expression, fusions, and splice variants. We integrated summarized Clinical Proteomic Tumor Analysis Consortium whole-cell proteomics and phospho-proteomics data and miRNA sequencing data, as well as developed a methylation array harmonization workflow to include m-values, beta-values, and copy number calls. OpenPedCan contains reproducible, dockerized workflows in GitHub, CAVATICA, and Amazon Web Services (AWS) to deliver harmonized and processed data from over 60 scalable modules, which can be leveraged both locally and on AWS. The processed data are released in a versioned manner and accessible through CAVATICA or AWS S3 download (from GitHub) and queryable through PedcBioPortal and the National Cancer Institute’s pediatric Molecular Targets Platform. Notably, we have expanded Pediatric Brain Tumor Atlas molecular subtyping to include methylation information to align with the World Health Organization 2021 Central Nervous System Tumor classifications, allowing us to create research-grade integrated diagnoses for these tumors. Conclusions: OpenPedCan data and its reproducible analysis module framework are openly available and can be utilized and/or adapted by researchers to accelerate discovery, validation, and clinical translation.

Background: Pediatric brain tumors are the leading cause of disease-related mortality in children, yet many aggressive tumors lack effective therapies. RNA splicing is a hallmark of cancer, but it has not yet been systematically studied in pediatric brain tumors. Methods: We analyzed 729 pediatric brain tumors spanning histologies and molecular subtypes to quantify differential tumor splicing. We developed the Splicing Burden Index (SBI) to enable cross-sample comparisons and performed hierarchical clustering of highly variable splice events to define splicing-informed tumor groups. These were integrated with clinical outcomes, pathway activity, and proteogenomic data. Recurrent splice events were prioritized for predicted functional impact, and in vitro perturbation studies were performed targeting the splicing kinase CDC-like kinase 1 (CLK1) . Results: SBI revealed substantial inter- and intra-histology heterogeneity. Clusters were enriched for histologies and molecular subtypes, several of which were independently associated with survival beyond histology and clinical covariates. Spliceosome pathway activity varied across clusters and was associated with worse survival, yet was not correlated with SBI, indicating distinct dimensions of splicing dysregulation. Functional prioritization identified a recurrent in CLK1 exon 4, required for canonical kinase activity. CLK1 exon 4 inclusion followed an oncofetal pattern and showed context-dependent associations with outcome distinct from total CLK1 expression. Pharmacologic inhibition and exon 4-specific perturbation of CLK1 reduced tumor cell viability and disrupted cancer-relevant splicing and transcriptional programs. Conclusions: This study systematically characterizes splicing in pediatric brain tumors, identifies splicing-informed subgroups, and prioritizes CLK1 exon 4 as an oncofetal tumor-specific event, motivating further preclinical exploration. Key Points: ● Splicing analysis of 729 pediatric CNS tumors identifies splicing-defined clusters. ● CLK1 exon 4 inclusion is widespread and developmentally regulated. ● Exon-level CLK1 regulation shows context-dependent links to prognosis in aggressive CNS tumors.

Background In 2019, the Open Pediatric Brain Tumor Atlas (OpenPBTA) was created as a global, collaborative open-science initiative to genomically characterize 1,074 pediatric brain tumors and 22 patient-derived cell lines. Here, we extend the OpenPBTA to create the Open Pediatric Cancer (OpenPedCan) Project, a harmonized open-source multi-omic dataset from 6,112 pediatric cancer patients with 7,096 tumor events across more than 100 histologies. Combined with RNA-Seq from the Genotype-Tissue Expression (GTEx) and The Cancer Genome Atlas (TCGA), OpenPedCan contains nearly 48,000 total biospecimens (24,002 tumor and 23,893 normal specimens). Findings We utilized Gabriella Miller Kids First (GMKF) workflows to harmonize WGS, WXS, RNA-seq, and Targeted Sequencing datasets to include somatic SNVs, InDels, CNVs, SVs, RNA expression, fusions, and splice variants. We integrated summarized CPTAC whole cell proteomics and phospho-proteomics data, miRNA-Seq data, and have developed a methylation array harmonization workflow to include m-values, beta-vales, and copy number calls. OpenPedCan contains reproducible, dockerized workflows in GitHub, CAVATICA, and Amazon Web Services (AWS) to deliver harmonized and processed data from over 60 scalable modules which can be leveraged both locally and on AWS. The processed data are released in a versioned manner and accessible through CAVATICA or AWS S3 download (from GitHub), and queryable through PedcBioPortal and the NCI’s pediatric Molecular Targets Platform. Notably, we have expanded PBTA molecular subtyping to include methylation information to align with the WHO 2021 Central Nervous System Tumor classifications, allowing us to create research-grade integrated diagnoses for these tumors. Conclusions OpenPedCan data and its reproducible analysis module framework are openly available and can be utilized and/or adapted by researchers to accelerate discovery, validation, and clinical translation.

Triple-negative breast cancer (TNBC) is an aggressive form of breast cancer associated with an early age of onset, greater propensity towards metastasis, and poorer clinical outcomes. It accounts for 10% to 20% of newly diagnosed breast cancer cases and disproportionately affects individuals from the African American race. While TNBC is sensitive to chemotherapy, it is also prone to relapse. This is because chemotherapy successfully targets the primary TNBC tumor cell but often fails to target the subpopulation of TNBC stem cells. TNBC stem cells display cancerous traits such as cell cycle progression, survival, proliferation, apoptosis inhibition, and epithelial-mesenchymal transition. To study the cancer initiating behavior of the TNBC stem cells, we studied their underlying signaling pathways using Boolean networks(BN). BNs are effective in capturing the causal interactions taking place in signaling pathways. We built the BN from the pathway literature and used it to evaluate the efficacies of eleven targeted inhibitory drugs in suppressing cancer-promoting genes. We simulated the BN when the pathways had single or multiple mutations, with a maximum of three mutations at a time. Our findings indicated that STAT3 , GLI , and NF-κB are the most optimal targets for inhibition. These genes are known regulators of the cancer-promoting genes in the pathway,hence our model agrees with the existing biological literature. Therefore inhibiting these three genes has the potential to prevent TNBC relapse. Additionally, our studies found that drug efficacies decreased as mutations increased in the pathway. Furthermore, we noticed that combinations of drugs performed better than single drugs.

Triple-negative breast cancer (TNBC) is an aggressive form of breast cancer associated with an early age of onset, greater propensity towards metastasis, and poorer clinical outcomes. It accounts for 10% to 20% of newly diagnosed breast cancer cases and disproportionately affects individuals from the African American race. While TNBC is sensitive to chemotherapy, it is also prone to relapse. This is because chemotherapy successfully targets the primary TNBC tumor cell but often fails to target the subpopulation of TNBC stem cells. TNBC stem cells display cancerous traits such as cell cycle progression, survival, proliferation, apoptosis inhibition, and epithelial-mesenchymal transition. To study the cancer initiating behavior of the TNBC stem cells, we studied their underlying signaling pathways using Boolean networks(BN). BNs are effective in capturing the causal interactions taking place in signaling pathways. We built the BN from the pathway literature and used it to evaluate the efficacies of eleven targeted inhibitory drugs in suppressing cancer-promoting genes. We simulated the BN when the pathways had single or multiple mutations, with a maximum of three mutations at a time. Our findings indicated that STAT3, GLI , and NF-κB are the most optimal targets for inhibition. These genes are known regulators of the cancer-promoting genes in the pathway,hence our model agrees with the existing biological literature. Therefore inhibiting these three genes has the potential to prevent TNBC relapse. Additionally, our studies found that drug efficacies decreased as mutations increased in the pathway. Furthermore, we noticed that combinations of drugs performed better than single drugs.

Drought is a natural hazard that affects crops by inducing water stress. Water stress, induced by drought accounts for more loss in crop yield than all the other causes combined. With the increasing frequency and intensity of droughts worldwide, it is essential to develop drought-resistant crops to ensure food security. In this paper, we model multiple drought signaling pathways in Arabidopsis using Bayesian networks to identify potential regulators of drought-responsive reporter genes. Genetically intervening at these regulators can help develop drought-resistant crops. We create the Bayesian network model from the biological literature and determine its parameters from publicly available data. We conduct inference on this model using a stochastic simulation technique known as likelihood weighting to determine the best regulators of drought-responsive reporter genes. Our analysis reveals that activating MYC2 or inhibiting ATAF1 are the best single node intervention strategies to regulate the drought-responsive reporter genes. Additionally, we observe simultaneously activating MYC2 and inhibiting ATAF1 is a better strategy. The Bayesian network model indicated that MYC2 and ATAF1 are possible regulators of the drought response. Validation experiments showed that ATAF1 negatively regulated the drought response. Thus intervening at ATAF1 has the potential to create drought-resistant crops.

Lysine is the first limiting essential amino acid in rice because it is present in the lowest quantity compared to all the other amino acids. Amino acids are the building block of proteins and play an essential role in maintaining the human body’s healthy functioning. Rice is a staple food for more than half of the global population; thus, increasing the lysine content in rice will help improve global health. In this paper, we studied the lysine biosynthesis pathway in rice (Oryza sativa) to identify the regulators of the lysine reporter gene LYSA (LOC_Os02g24354). Genetically intervening at the regulators has the potential to increase the overall lysine content in rice. We modeled the lysine biosynthesis pathway in rice seedlings under normal and saline (NaCl) stress conditions using Bayesian networks. We estimated the model parameters using experimental data and identified the gene DAPF(LOC_Os12g37960) as a positive regulator of the lysine reporter gene LYSA under both normal and saline stress conditions. Based on this analysis, we conclude that the gene DAPF is a potent candidate for genetic intervention. Upregulating DAPF using methods such as CRISPR-Cas9 gene editing strategy has the potential to upregulate the lysine reporter gene LYSA and increase the overall lysine content in rice.

Drought is a natural hazard that affects crops by inducing water stress. Water stress, induced by drought, accounts for more loss in crop yield than all the other causes combined. With the increasing frequency and intensity of droughts worldwide, it is essential to develop drought-resistant crops to ensure food security. In this paper, we model multiple drought signaling pathways in Arabidopsis using Bayesian networks to identify potential regulators of drought-responsive reporter genes. Genetically intervening at these regulators can help develop drought-resistant crops. We create the Bayesian network model from the biological literature and determine its parameters from publicly available data. We conduct inference on this model using a stochastic simulation technique known as likelihood weighting to determine the best regulators of drought-responsive reporter genes. Our analysis reveals that activating MYC2 or inhibiting ATAF1 are the best single node intervention strategies to regulate the drought-responsive reporter genes. Additionally, we observe simultaneously activating MYC2 and inhibiting ATAF1 is a better strategy. The Bayesian network model indicated that MYC2 and ATAF1 are possible regulators of the drought response. Validation experiments showed that ATAF1 negatively regulated the drought response. Thus intervening at ATAF1 has the potential to create drought-resistant crops.