Publications by authors named "Shilpa N Kobren"
Article Synopsis
- Rare genetic conditions affect about 1 in 17 people globally, making it challenging to identify specific variants that cause these diseases, especially in undiagnosed cases.
- Clinicians often rely on variant pathogenicity predictions to differentiate harmful genetic variants from benign ones, but these methods struggle with complex cases, necessitating extensive manual analysis.
- The introduction of VarPPUD offers a more accurate tool, achieving 79.3% accuracy and 77.5% precision in identifying pathogenic variants from difficult cases, outperforming traditional methods and allowing for deeper analysis of genetic factors.
Article Synopsis
- The study explores complex biological mechanisms behind extreme symptoms in rare disease patients, emphasizing the need for treatments targeting underlying causes rather than just symptoms.
- It focuses on seizures as a common symptom in patients with ultrarare disorders and analyzes genotype and phenotype data from the UK Biobank to uncover related biological pathways.
- The researchers present case studies of undiagnosed patients with seizures and discuss how their findings can provide insights into the molecular mechanisms of rare diseases, highlighting the importance of large-scale data analysis in understanding these conditions.
Article Synopsis
- Recent advancements in genomics for diagnosing rare diseases focus on "N-of-1" analyses, allowing for tailored studies on individual patients with ultra-rare conditions.
- The Undiagnosed Diseases Network (UDN) enables collaborative research across various U.S. clinical and research centers, which enhances the ability to analyze whole genome sequencing data from multiple patients simultaneously.
- Introducing a new software package, RaMeDiES, the team provides tools for automated comparisons of genomic data, leading to novel disease associations and improving overall understanding of genetic links to these rare diseases.
Article Synopsis
- Expansions of tandem repeats (TRs) are linked to about 60 genetic diseases, and finding more pathogenic repeats could improve disease diagnosis.
- RExPRT (Repeat EXpansion Pathogenicity pRediction Tool) is a machine learning tool designed to differentiate harmful TR expansions from harmless ones.
- The tool has shown impressive results, achieving an average precision of 93% and recall of 83%, making it helpful for prioritizing which genetic candidates to study further in large-scale research.
Article Synopsis
- Rare Mendelian disorders affect 300-400 million people globally and present significant diagnostic challenges due to the complexity and lack of data.
- Existing automated tools struggle to identify causal genes for these disorders as there are limited datasets with unpublished cases for evaluation.
- The authors developed a new computational pipeline that simulates clinical datasets to create realistic patient profiles, allowing researchers to test and improve gene prioritization methods for diagnosing novel genetic conditions.
Am J Med Genet A
October 2023
Article Synopsis
- - The study explores the role of mosaicism in genetic diseases and how it relates to disease-causing variants that arise spontaneously (de novo variants) in families, involving data from nearly 15,000 individuals.
- - Researchers found that about 4.51% of individuals with confirmed genetic diseases showed mosaic genetic disease (MGD), and approximately 2.86% of parents had parental mosaicism, especially in cases involving de novo variants.
- - The findings highlight the complexity and variability of MGD, suggesting it contributes significantly to genetic disorders, although further investigation is needed to better understand its implications for diagnoses and familial risks.
Article Synopsis
- Genomic sequencing is a crucial tool for identifying genetic issues in rare Mendelian diseases, and while techniques are improving, there’s a need to find common practices and areas for better methods in the diagnostic process.*
-
- A study involved gathering information from a genetic testing lab and 11 clinical sites across the U.S. to understand the computational strategies used in workflow for analyzing genomic data.*
-
- Results showed that while there are solid methods for initial data processing, significant variation exists in later steps, especially in how data is prioritized and integrated, indicating that enhancing detection of certain variants could aid in resolving tough undiagnosed cases.*
Article Synopsis
- A key issue in cancer genomics involves identifying genes that contribute to cancer and understanding how they work.
- The authors propose a new framework that detects cancer-related genes by analyzing somatic mutations in interaction sites across various tumors.
- Their software, PertInInt, processes extensive data to reveal both existing and new cancer genes and shows that many mutations affect key interaction sites, highlighting interaction disruption as a common factor in cancer development.
Nucleic Acids Res
January 2019
Article Synopsis
- Domains are critical units in proteins that interact with DNA, RNA, and other molecules, but their various interaction modes haven't been thoroughly analyzed.
- The study introduces a method that tracks how often specific positions within protein domains interact with different ligands, using data from ~91,000 protein co-complex structures.
- The findings, compiled in a resource called the InteracDome, can help identify potential ligand-binding sites in 2,152 domains and link these sites to diseases, highlighting their relevance in mutations associated with genetic disorders and cancer.
