A5013325970- CRISPR and Genetic Engineering
- RNA and protein synthesis mechanisms
- Bioinformatics and Genomic Networks
- Bayesian Modeling and Causal Inference
- Advanced Causal Inference Techniques
- Amyotrophic Lateral Sclerosis Research
- Machine Learning in Healthcare
- Chromosomal and Genetic Variations
- Imbalanced Data Classification Techniques
- Cytomegalovirus and herpesvirus research
- Microbial Metabolic Engineering and Bioproduction
- RNA modifications and cancer
- Colorectal Cancer Screening and Detection
- Advanced biosensing and bioanalysis techniques
- COVID-19 diagnosis using AI
- Histone Deacetylase Inhibitors Research
- Additive Manufacturing Materials and Processes
- Welding Techniques and Residual Stresses
- Advanced Malware Detection Techniques
- RNA Research and Splicing
- Virus-based gene therapy research
- Machine Learning in Materials Science
- Innovation and Socioeconomic Development
- Fungal and yeast genetics research
- HIV Research and Treatment
Massachusetts Institute of Technology
2017-2025
University of Toronto
2025
McGovern Institute for Brain Research
2021-2025
NUS High School of Mathematics and Science
2023
University of Sheffield
2023
Broad Institute
2017-2021
Massachusetts General Hospital
2020
Programmable RNA-guided DNA nucleases perform numerous roles in prokaryotes, but the extent of their spread outside prokaryotes is unclear. Fanzors, eukaryotic homolog prokaryotic TnpB proteins, have been detected genomes eukaryotes and large viruses, activity functions remain unknown. Here, we characterize Fanzors as RNA-programmable endonucleases, using biochemical cellular evidence. We found diverse that frequently associate with various transposases. Reconstruction evolution revealed...
Abstract Programmable and multiplexed genome integration of large, diverse DNA cargo independent repair remains an unsolved challenge editing. Current gene approaches require double-strand breaks that evoke damage responses rely on pathways are inactive in terminally differentiated cells. Furthermore, CRISPR-based bypass double stranded breaks, such as Prime editing, limited to modification or insertion short sequences. We present Addition via Site-specific Targeting Elements, PASTE, which...
Abstract Combining genetic and cell-type-specific proteomic datasets can generate biological insights therapeutic hypotheses, but a technical statistical framework for such analyses is lacking. Here, we present an open-source computational tool called Genoppi (lagelab.org/genoppi) that enables robust, standardized, intuitive integration of quantitative results with data. We use to analyze 16 protein interaction four proteins (BCL2, TDP-43, MDM2, PTEN) involved in cancer neurological disease....
Healthcare providers are increasingly using machine learning to predict patient outcomes make meaningful interventions. However, despite innovations in this area, deep models often struggle match performance of shallow linear predicting these outcomes, making it difficult leverage such techniques practice. In work, motivated by the task clinical prediction from insurance claims, we present a new technique called reverse distillation which pretrains high-performing for initialization. We use...
TnpB proteins are RNA-guided nucleases that broadly associated with IS200/605 family transposons in prokaryotes. homologs, named Fanzors, have been detected genomes of some eukaryotes and large viruses, but their activity functions remain unknown. We searched diverse viruses for homologs identified numerous putative often various transposases, suggesting they encoded mobile genetic elements. Reconstruction the evolution these nucleases, which we rename Horizontally-transferred Eukaryotic...
The presence of undesirable microstructural features in additively manufactured components, such as cracks, pores and lack fusion defects presents a challenge for engineers, particularly if these components are applied structure-critical applications. Such might need to be manually classified, counted their size distributions measured during metallographic evaluation, which is time-consuming task. In this study, the performance two supervised machine learning methods ( kth-nearest neighbours...
Dynamic malware analysis, which has been a major field in analysis and detection, involves executing controlled environment observing its behavior. reports include Windows API calls, are extracted as data source for statistical features, have allowed effective detection. However, existing works neglect certain critical information about the calls when constructing feature vectors. In this work, we develop novel vector, taking into account not only name arguments but also other features such...
Abstract Combining genetic and cell-type-specific proteomic datasets can lead to new biological insights therapeutic hypotheses, but a technical statistical framework for such analyses is lacking. Here, we present an open-source computational tool called Genoppi that enables robust, standardized, intuitive integration of quantitative results with data. We used analyze sixteen protein interaction four proteins (TDP-43, MDM2, PTEN, BCL2) involved in cancer neurological disease. Through...
Abstract Summary Integrating protein-protein interaction experiments and genetic datasets can lead to new insight into the cellular processes implicated in diseases, but this integration is technically challenging. Here, we present Genoppi, a web application that integrates quantitative proteomics data results from genome-wide association studies or exome sequencing projects, highlight biological relationships might otherwise be difficult discern. Written R, Python Bash script, Genoppi...