Kaiyuan Zhu

Cycle-Extractor accurately reconstructs ecDNA structures from both short- and long-read sequencing data, outperforming existing methods in speed and completeness.

The paper introduces ec3D, a method to reconstruct 3D structures of ecDNA from Hi-C data, revealing spherical configurations and unique long-range interactions in cancer.

TX-Phase securely phases private genomes using Trusted Execution Environments, achieving high accuracy and efficiency on UK Biobank and Haplotype Reference Consortium datasets.

The paper improves the SkSES method for privacy-preserving GWAS by integrating a learning-augmented approach, achieving up to 40% higher accuracy under memory constraints.

The paper introduces OM2BFB, an algorithm detecting BFB amplifications using OGMs, showing high precision and recall, and finds BFB events are common in certain cancers but not linked to intratumoral heterogeneity or rapid genome change.

CoRAL improves ecDNA structure reconstruction using long-read sequencing, outperforming short-read methods in simulations and cell-line datasets.

CAMMiQ improves microbial strain-level detection and quantification in single-cell metagenomics by using combinatorial optimization and doubly-unique substrings, outperforming existing methods in accuracy.

This paper introduces a privacy-preserving genotype imputation method using a Trusted Execution Environment to protect genetic data while maintaining accuracy comparable to existing tools.

SkSES enables secure and privacy-preserving SNP analysis across encrypted genomic data using sketching algorithms in Intel SGX enclaves, demonstrating efficient SNP querying in a large dataset.

The paper introduces cd-CAP, a method to detect conserved alteration patterns in cancer subnetworks across tumor samples.