# Xiaole Shirley Liu
Xiaole Shirley Liu, publishing as X. Shirley Liu, is a computational biologist known for building widely used algorithms for ChIP-seq (DNA-sequencing method mapping where proteins bind the genome) analysis, [CRISPR screen](https://www.edgepedia.org/crispr-screen) statistics, and tumor immune characterization. She was Professor of Biostatistics and Computational Biology at Dana-Farber Cancer Institute and Harvard University until 2022.[1](https://liulab-dfci.github.io/people/shirley_liu)
| | |
|---|---|
| **Field** | Computational biology and bioinformatics; cancer genomics and immunooncology[1](https://liulab-dfci.github.io/people/shirley_liu) |
| **Education** | Ph.D. in Biomedical Informatics and Ph.D. minor in Computer Science, Stanford University, 2002[1](https://liulab-dfci.github.io/people/shirley_liu) |
| **Academic posts** | Professor of Biostatistics and Computational Biology, Dana-Farber/Harvard; Director of the Center of Functional Cancer Epigenetics, Dana-Farber, until 2022[1](https://liulab-dfci.github.io/people/shirley_liu) |
| **Signature work** | MACS (Genome Biology, 2008); TIDE immunotherapy-response biomarker[3](https://link.springer.com/article/10.1186/gb-2008-9-9-r137) • [4](https://liulab-dfci.github.io/software/) |
| **Major honors** | Sloan Research Fellowship (2008); ISCB Fellow (2019); ISCB Innovator Award and Benjamin Franklin Award (2020); AIMBE College of Fellows (2022)[5](https://cqb.pku.edu.cn/cqben/info/1040/1871.htm) • [2](https://gv20tx.com/news/details/20220225) |
| **Selected leadership** | Principal investigator of the NCI Cancer Moonshot Cancer Immune Data Commons[1](https://liulab-dfci.github.io/people/shirley_liu) |
## Education and early career
Liu received a Ph.D. in Biomedical Informatics and a Ph.D. minor in Computer Science from Stanford University in 2002.[1](https://liulab-dfci.github.io/people/shirley_liu) As a member of the mod/ENCODE consortium she helped establish best practices in ChIP-chip and ChIP-seq experiments, and with colleagues she generated the first high-throughput nucleosome map in the human genome and identified the chromatin signature of embryonic pluripotency.[1](https://liulab-dfci.github.io/people/shirley_liu) The International Society for Computational Biology has credited her as a pioneer in using chromatin dynamics to predict the trans-factors and cis-elements involved in biological processes and disease.[6](https://www.iscb.org/ismb2020/whats-happening/distinguished-keynotes/xiaole-shirley-liu)
## Career at Dana-Farber and Harvard
She was Professor of Biostatistics and Computational Biology at Harvard University and Director of the Center of Functional Cancer Epigenetics at Dana-Farber Cancer Institute until 2022.[1](https://liulab-dfci.github.io/people/shirley_liu) Her laboratory developed widely used algorithms for transcription factor motif finding, ChIP-chip, and ChIP-seq, chromatin accessibility, CRISPR screen analysis, and tumor immune characterization.[1](https://liulab-dfci.github.io/people/shirley_liu) She also served as principal investigator of the Cancer Immune Data Commons, a National Cancer Institute Cancer Moonshot project aimed at identifying biomarkers for optimizing cancer immunotherapy strategies.[1](https://liulab-dfci.github.io/people/shirley_liu) Her group's work applied epigenomic and computational analysis to hormone receptor therapies, epigenetic inhibitors, gamma-secretase inhibitors, receptor tyrosine kinase inhibitors, and immune checkpoint inhibitors across cancer types.[2](https://gv20tx.com/news/details/20220225)
## Representative work
**MACS** (Model-based Analysis of ChIP-Seq), published in *Genome Biology* in 2008, analyzes short-read ChIP-seq data by empirically modeling the shift size of ChIP-seq tags, using that model to improve the spatial resolution of predicted binding sites, and it applies a dynamic [Poisson distribution](https://www.edgepedia.org/poisson-distribution) to capture local biases in the genome for more robust peak predictions.[3](https://link.springer.com/article/10.1186/gb-2008-9-9-r137) In the original paper it was compared against ChIPSeq Peak Finder, FindPeaks, and QuEST, and for FoxA1 and NRSF datasets it consistently produced fewer false positives than the other three methods.[3](https://link.springer.com/article/10.1186/gb-2008-9-9-r137) A later benchmark surveying 30 peak-calling methods on 300 simulated and 3 real datasets found that BCP and MACS2 had the best operating characteristics on simulated transcription factor binding data, and that Poisson-based significance testing, as MACS2 uses, is more powerful than Binomial testing.[7](https://pmc.ncbi.nlm.nih.gov/articles/PMC5429005/) According to the MACS3 documentation, the tool has continued as MACS3, maintained to address robust analysis of genome-wide protein-DNA interactions.[8](https://macs3-project.github.io/MACS/)
Her immune-oncology tools address a different problem: quantifying the immune landscape of a tumor from bulk sequencing. TIMER (Tumor IMmune Estimation Resource) estimates the clinical relevance of tumor infiltrates of B cells, CD4 and CD8 T cells, neutrophils, macrophages, and dendritic cells; TIDE is a gene expression biomarker for predicting clinical response to immune checkpoint blockade; and TRUST4 reconstructs TCR/BCR immune repertoires from bulk or single-cell tumor RNA-seq data.[4](https://liulab-dfci.github.io/software/) The lab's toolset also includes TIMER2, DeepBCR, TISMO, RIMA, and MAGeCKFlute.[4](https://liulab-dfci.github.io/software/)
## GV20 Therapeutics
According to a Nature profile of GV20 Therapeutics, Liu co-founded the company in 2016 and became its CEO on February 1, 2022, leaving Dana-Farber and Harvard after nearly two decades on the faculty.[9](https://www.nature.com/articles/d43747-023-00067-3.pdf) Its in-house STEAD platform (Simultaneous Target Evaluation and Antibody Discovery) integrates AI, genomics, and disease biology to find targets and create therapies.[10](https://gv20tx.com/news/details/20240408)
The lead program, GV20-0251, targets IGSF8, which GV20's modeling identified as a novel innate immune checkpoint highly expressed on malignant cells with antigen presentation defects; the company presented preclinical data at the AACR Annual Meeting in April 2024, describing the antibody as the first AI-designed antibody against an AI-predicted target to advance into the clinic, and the ongoing Phase I study in solid tumors is registered as NCT05669430.[10](https://gv20tx.com/news/details/20240408) In September 2024 the program was described as a first-in-class, fully human, Fc-attenuated IgG1 monoclonal antibody that showed favorable safety and promising monotherapy efficacy in the multi-center Phase 1/2 trial in advanced solid tumors.[11](https://biotochina.org/2024/09/24/gv20-therapeutics-to-speak-at-access-china-partnering-forum/) Preclinical work showed anti-IGSF8 antibodies enhance NK cell cytotoxicity, dendritic cell antigen presentation, and [T cell](https://www.edgepedia.org/t-cell) activation, both alone and in combination with anti-PD1.[10](https://gv20tx.com/news/details/20240408)
## Awards and honors
Liu received the Sloan Research Fellowship in 2008, the Endocrine Society's Weitzman Outstanding Early Career Investigator Award in 2016, and became a Breast Cancer Research Foundation Investigator in 2017.[5](https://cqb.pku.edu.cn/cqben/info/1040/1871.htm) • [2](https://gv20tx.com/news/details/20220225) She became a Fellow of ISCB in 2019, and in 2020 received both the ISCB Innovator Award, given to a scientist within two decades of the Ph.D. who has made outstanding contributions and continues to forge new directions, and the Benjamin Franklin Award for Open Access in the Life Sciences.[6](https://www.iscb.org/ismb2020/whats-happening/distinguished-keynotes/xiaole-shirley-liu) • [2](https://gv20tx.com/news/details/20220225)
## Recent developments
GV20-0251 moved through its Phase 1/2 study with reports of early clinical activity in 2024, and the company describes a pipeline including ADCs and degraders.[10](https://gv20tx.com/news/details/20240408) • [11](https://biotochina.org/2024/09/24/gv20-therapeutics-to-speak-at-access-china-partnering-forum/) Her older tools remain in active use and under comparison: a November 2024 preprint benchmarking MACS2 against SEACR, GoPeaks, and LanceOtron on CUT&RUN data found substantial variability in peak calling efficacy, with each method showing distinct strengths in sensitivity, precision, and applicability depending on the histone mark, a reminder that the right peak caller depends on the assay.[12](https://www.biorxiv.org/content/10.1101/2024.11.13.622880v1)
## References
1. [shirley_liu, Liu Lab](https://liulab-dfci.github.io/people/shirley_liu)
2. [GV20 Therapeutics, News: Dr. Xiaole Shirley Liu selected to AIMBE 2022 College of Fellows](https://gv20tx.com/news/details/20220225)
3. [Model-based Analysis of ChIP-Seq (MACS), Genome Biology](https://link.springer.com/article/10.1186/gb-2008-9-9-r137)
4. [Software, Liu Lab](https://liulab-dfci.github.io/software/)
5. [Integrated Genomics and Computation for Cancer Target and Drug Discovery, Center for Quantitative Biology, Peking University](https://cqb.pku.edu.cn/cqben/info/1040/1871.htm)
6. [Xiaole Shirley Liu, ISMB 2020 Distinguished Keynote, ISCB](https://www.iscb.org/ismb2020/whats-happening/distinguished-keynotes/xiaole-shirley-liu)
7. [Features that define the best ChIP-seq peak calling algorithms, PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC5429005/)
8. [MACS: Model-based Analysis for ChIP-Seq, MACS3 documentation](https://macs3-project.github.io/MACS/)
9. [Nature portfolio profile of GV20 Therapeutics](https://www.nature.com/articles/d43747-023-00067-3.pdf)
10. [GV20 Therapeutics, News (April 8, 2024): AACR presentation on GV20-0251](https://gv20tx.com/news/details/20240408)
11. [GV20 Therapeutics to Speak at ACCESS CHINA Partnering Forum, BioToChina.org](https://biotochina.org/2024/09/24/gv20-therapeutics-to-speak-at-access-china-partnering-forum/)
12. [Benchmarking Peak Calling Methods for CUT&RUN, bioRxiv](https://www.biorxiv.org/content/10.1101/2024.11.13.622880v1)
---
*Topic: Encyclopedia › Life and health › Life and health scientists › Medical and health researchers › Researchers in molecular diagnostics, pathology, medical imaging, and precision medicine › Bioinformatics and multi-omics integration*
*Initially written Sep 20, 2026 · Reviewed: Sep 25, 2026 · Edited: Sep 25, 2026; Oct 15, 2026 · Last review: Sep 25, 2026*
*Copyright 2026 EdgeChat AI, a subsidiary of Biostate AI.*
License: Edgepedia Community License 1.0, https://www.edgepedia.org/edgepedia/license