Hanaé Carrié
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Hanaé Carrié has a total of 40 co-authors. Here are the top co-authors by number of shared publications:
- Anders Jacobsen Skanderup
Genome Institute of Singapore, Agency for Science, Technology and Research, Singapore
3 shared publications Anna Gan
Genome Institute of Singapore (GIS), Agency for Science, Technology, and Research (A*STAR), Singapore, Singapore
3 shared publicationsNgak Leng Sim
Genome Institute of Singapore (GIS), Agency for Science, Technology, and Research (A*STAR), Singapore, Singapore
3 shared publicationsSaranya Thangaraju
Genome Institute of Singapore (GIS), Agency for Science, Technology, and Research (A*STAR), Singapore, Singapore
3 shared publications- Limsoon Wong
Department of Computer Science, School of Computing, National University of Singapore, 117417, Singapore
2 shared publications Polly Poon
Genome Institute of Singapore (GIS), Agency for Science, Technology and Research (A*STAR), Singapore, Singapore
2 shared publications
Latest Publications
- Open Access
- 18 December 2025
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Comprehensive benchmarking of methods for mutation calling in circulating tumor DNA
Detection of somatic mutations in cell-free DNA (cfDNA) is challenging due to low variant allele frequencies and extensive DNA degradation. Here we develop a benchmarking strategy using longitudinal patient-matched cfDNA samples from individuals with colorectal and breast cancer. Samples with high and ultra-low levels of tumor-derived DNA are combined into controlled dilution series that preserve the properties of authentic cell-free DNA, including each patient’s germline and blood-cell mutation backgrounds. Using deep whole-genome (150x) and exome (2,000x) sequencing, we define a reference set of ~37,000 single nucleotide variants and ~58,000 indels to benchmark nine somatic variant callers across varying ctDNA levels and sequencing depths. We also explore machine learning–based tuning of individual callers and identify features that improve accuracy in cfDNA. This benchmarking resource clarifies the detection limits of current approaches and provides practical guidance for selecting somatic variant calling methods in liquid biopsy applications.
- 18 December 2025
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Open AccessComprehensive benchmarking of methods for mutation calling in circulating tumor DNA
Detection of somatic mutations in cell-free DNA (cfDNA) is challenging due to low variant allele frequencies and extensive DNA degradation. Here we develop a benchmarking strategy using longitudinal patient-matched cfDNA samples from individuals with colorectal and breast cancer. Samples with high and ultra-low levels of tumor-derived DNA are combined into controlled dilution series that preserve the properties of authentic cell-free DNA, including each patient’s germline and blood-cell mutation backgrounds. Using deep whole-genome (150x) and exome (2,000x) sequencing, we define a reference set of ~37,000 single nucleotide variants and ~58,000 indels to benchmark nine somatic variant callers across varying ctDNA levels and sequencing depths. We also explore machine learning–based tuning of individual callers and identify features that improve accuracy in cfDNA. This benchmarking resource clarifies the detection limits of current approaches and provides practical guidance for selecting somatic variant calling methods in liquid biopsy applications.
- 7 March 2025
- Published by Springer Nature in Nature Biomedical Engineering
- Vol. 9 (3) , 307-319
A deep-learning model for quantifying circulating tumour DNA from the density distribution of DNA-fragment lengths
The quantification of circulating tumour DNA (ctDNA) in blood enables non-invasive surveillance of cancer progression. Here we show that a deep-learning model can accurately quantify ctDNA from the density distribution of cell-free DNA-fragment lengths. We validated the model, which we named ‘Fragle’, by using low-pass whole-genome-sequencing data from multiple cancer types and healthy control cohorts. In independent cohorts, Fragle outperformed tumour-naive methods, achieving higher accuracy and lower detection limits. We also show that Fragle is compatible with targeted sequencing data. In plasma samples from patients with colorectal cancer, longitudinal analysis with Fragle revealed strong concordance between ctDNA dynamics and treatment responses. In patients with resected lung cancer, Fragle outperformed a tumour-naive gene panel in the prediction of minimal residual disease for risk stratification. The method’s versatility, speed and accuracy for ctDNA quantification suggest that it may have broad clinical utility.
- 7 March 2025
- Published by Springer Nature in Nature Biomedical Engineering
- Vol. 9 (3) , 307-319
A deep-learning model for quantifying circulating tumour DNA from the density distribution of DNA-fragment lengths
The quantification of circulating tumour DNA (ctDNA) in blood enables non-invasive surveillance of cancer progression. Here we show that a deep-learning model can accurately quantify ctDNA from the density distribution of cell-free DNA-fragment lengths. We validated the model, which we named ‘Fragle’, by using low-pass whole-genome-sequencing data from multiple cancer types and healthy control cohorts. In independent cohorts, Fragle outperformed tumour-naive methods, achieving higher accuracy and lower detection limits. We also show that Fragle is compatible with targeted sequencing data. In plasma samples from patients with colorectal cancer, longitudinal analysis with Fragle revealed strong concordance between ctDNA dynamics and treatment responses. In patients with resected lung cancer, Fragle outperformed a tumour-naive gene panel in the prediction of minimal residual disease for risk stratification. The method’s versatility, speed and accuracy for ctDNA quantification suggest that it may have broad clinical utility.
- Open Access
Comprehensive benchmarking of methods for mutation calling in circulating tumor DNA
Detection of somatic mutations in cell-free DNA (cfDNA) is challenging due to low variant allele frequencies and pronounced DNA degradation. Here, we present a novel approach and resource for benchmarking of somatic variant calling algorithms in cfDNA samples from cancer patients. Using longitudinally collected cfDNA samples from colorectal and breast cancer patients, we identify patient-matched samples with high and ultra-low circulating tumor DNA (ctDNA) levels. These sample pairs, preserving patient-specific germline and somatic haematopoiesis variant backgrounds, were used to generate dilution series capturing characteristics of bona-fide cfDNA samples. To benchmark the accuracy and limit of detection of 9 somatic variant calling algorithms, we used deep Whole Genome Sequencing (WGS, 150x) and ultra-deep Whole Exome Sequencing (WES, 2,000x) to construct a reference set of ∼37,000 Single Nucleotide Variants and ∼58,000 Insertions/Deletions. We tested methods under variable ctDNA levels and depth of sequencing, generating guidelines for method choice depending on use case. Using a machine learning approach, we further evaluated the potential of fine-tuning individual variant callers, revealing features that may improve accuracy in cfDNA samples. Overall, we present a new resource for benchmarking of somatic variant calling methods in cfDNA, providing insights on method choice to realize the potential of liquid biopsies in precision oncology.
Open AccessComprehensive benchmarking of methods for mutation calling in circulating tumor DNA
Detection of somatic mutations in cell-free DNA (cfDNA) is challenging due to low variant allele frequencies and pronounced DNA degradation. Here, we present a novel approach and resource for benchmarking of somatic variant calling algorithms in cfDNA samples from cancer patients. Using longitudinally collected cfDNA samples from colorectal and breast cancer patients, we identify patient-matched samples with high and ultra-low circulating tumor DNA (ctDNA) levels. These sample pairs, preserving patient-specific germline and somatic haematopoiesis variant backgrounds, were used to generate dilution series capturing characteristics of bona-fide cfDNA samples. To benchmark the accuracy and limit of detection of 9 somatic variant calling algorithms, we used deep Whole Genome Sequencing (WGS, 150x) and ultra-deep Whole Exome Sequencing (WES, 2,000x) to construct a reference set of ∼37,000 Single Nucleotide Variants and ∼58,000 Insertions/Deletions. We tested methods under variable ctDNA levels and depth of sequencing, generating guidelines for method choice depending on use case. Using a machine learning approach, we further evaluated the potential of fine-tuning individual variant callers, revealing features that may improve accuracy in cfDNA samples. Overall, we present a new resource for benchmarking of somatic variant calling methods in cfDNA, providing insights on method choice to realize the potential of liquid biopsies in precision oncology.