Steven Gareth Williams

Steven Gareth Williams

Auto-generated profile, based on publicly available information

Steven Gareth Williams

ProCan, Children’s Medical Research Institute, Faculty of Medicine and Health, The University of Sydney, Westmead, Australia

Research Focus & Metrics

Academic Career Length

13Years (2014-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
30 July 2020 in Nature Communications
Verified
Journal Article
Open access
PDF file
RPRebecca Poulos
Rebecca Poulos
PHPeter Hains
Peter Hains
RSRohan Shah
Rohan Shah
NLNatasha Lucas
Natasha Lucas
DXDylan Xavier
Dylan Xavier
QZQing Zhong
Qing Zhong
Reproducible research is the bedrock of experimental science. To enable the deployment of large-scale proteomics, we assess the reproducibility of mass spectrometry (MS) over time and across instruments and develop computational methods for improving quantitative accuracy. We perform 1560 data independent acquisition (DIA)-MS runs of eight samples containing known proportions of ovarian and prostate cancer tissue and yeast, or control HEK293T cells. Replicates are run on six mass spectrometers operating continuously with varying maintenance schedules over four months, interspersed with ~5000 other runs. We utilise negative controls and replicates to remove unwanted variation and enhance biological signal, outperforming existing methods. We also design a method for reducing missing values. Integrating these computational modules into a pipeline (ProNorM), we mitigate variation among instruments over time and accurately predict tissue proportions. We demonstrate how to improve the quantitative analysis of large-scale DIA-MS data, providing a pathway toward clinical proteomics.

9 June 2025 in Cancer Discovery
Verified
Journal Article
Open access
ZCZhaoxiang Cai
Zhaoxiang Cai
EBEmma Boys
Emma Boys
ZNZainab Noor
Zainab Noor
AAAdel T. Aref
Adel T. Aref
DXDylan Xavier
Dylan Xavier
RRRoger Reddel
Roger Reddel
Artificial intelligence applications in biomedicine face major challenges from data privacy requirements. To address this issue for clinically annotated tissue proteomic data, we developed a federated deep learning approach (ProCanFDL), training local models on simulated sites containing data from a pan-cancer cohort (n = 1,260) and 29 cohorts held behind private firewalls (n = 6,265), representing 19,930 replicate data-independent acquisition mass spectrometry runs. Local parameter updates were aggregated to build the global model, achieving a 43% performance gain on the hold-out test set (n = 625) in 14 cancer subtyping tasks compared with local models and matching centralized model performance. The approach’s generalizability was demonstrated by retraining the global model with data from two external, data-independent acquisition mass spectrometry cohorts (n = 55) and eight acquired by tandem mass tag proteomics (n = 832). ProCanFDL presents a solution for internationally collaborative machine learning initiatives using proteomic data, for example, for discovering predictive biomarkers or treatment targets while maintaining data privacy. Significance: A federated deep learning approach applied to human proteomic data, acquired using two distinct proteomic technologies from 40 tumor cohorts across eight countries, enabled accurate cancer histopathologic subtyping while preserving data privacy. This approach will enable the privacy-compliant development of large-scale proteomic artificial intelligence models, including foundation models, across institutions globally.

13 March 2014 in Monatshefte für Chemie - Chemical Monthly
Verified
Journal Article
Closed access
PDF file
Alison Thavary UngAlison Thavary Ung
Alison Thavary Ung
SWSteven G. Williams
Steven G. Williams
AAAlexander Angeloski
Alexander Angeloski
JAJason Ashmore
Jason Ashmore
UKUnnikrishnan Kuzhiumparambil
Unnikrishnan Kuzhiumparambil
RBRoger Bishop
Roger Bishop

1 January 2017 in Tetrahedron
Verified
Journal Article
Closed access
SWSteven G. Williams
Steven G. Williams
MBMohan Bhadbhade
Mohan Bhadbhade
RBRoger Bishop
Roger Bishop
AUAlison T. Ung
Alison T. Ung

1 October 2025 in Cancer Research Communications
Verified
Journal Article
Open access
AAAdel T Aref
Adel T Aref
JGJason Grealey
Jason Grealey
MPMohashin Pathan
Mohashin Pathan
ZNZainab Noor
Zainab Noor
AAAsim Anees
Asim Anees
AGAnthony J Gill
Anthony J Gill
Pancreatic ductal adenocarcinoma (PDA) is an aggressive malignancy that lacks reliable biomarkers to guide treatment decisions. Effective prognostic tools are needed to improve its clinical management. We conducted a comprehensive proteomic analysis on 115 PDA patient samples with matched adjacent normal tissue. A 20-protein diagnostic panel was identified (LGALS1, ANXA2, LGALS3BP, CTSD, S100P, COL12A1, SFN, THBS2, CTHRC1, THBS1, SERPINB5, LAMC2, POSTN, CEACAM6, CTSE, PLEC, PKM, S100A11, TAGLN2, ALDOA). Consensus clustering analysis identified four prognostic proteomic subtypes. Subtypes with poorer prognoses exhibited upregulation of neutrophil degranulation, extracellular matrix remodeling, focal adhesion, Mesenchymal Epithelial Transition, collagen formation, and PI3K-Akt-mTOR-related pathways, indicating a predominance of basal-like and activated stromal features. In tumors with homologous recombination deficiency or Catalogue of Somatic Mutations in Cancer Signature-3, several immune-related proteins were enriched. An 18-protein (PURB, SDCBP2, CD2BP2, GALM, SERPINA3, OAS3, FAN1, ZPR1, KRT2, NUDT2, SMNDC1, SERPINA4, CUTA, WDR36, POSTN, CLEC11A, PEX14, and PI4KA) risk score was developed and validated using multicox regression analyses with LASSO regularization. The risk score demonstrated independent prognostic significance for overall survival and recurrence, and was validated in an independent proteomic dataset generated using a different proteomic technology. This study thus introduces four novel prognostic PDA subtypes, and an 18-protein risk score validated in an independent dataset, which shows promise for improving survival prediction and could serve as a valuable tool for personalized treatment guidance. Significance: The findings from this study have significant implications for the future of pancreatic cancer management. By identifying a 20-protein panel with diagnostic and screening potential, this research provides a foundation for developing early detection tools for PDA, an aggressive cancer with limited treatment options. The classification of PDA into four proteomic subtypes with distinct prognostic outcomes paves the way for subtype-specific therapeutic approaches, allowing clinicians to better stratify patients based on their risk profiles. Additionally, the validated 18-protein risk score, which enhances survival prediction and operates independently of existing clinical variables, represents a promising tool for personalized prognostic assessments. Incorporating these proteomic-based biomarkers into clinical practice could improve diagnostic accuracy, guide individualized treatment decisions, and ultimately enhance patient outcomes in PDA. This study underscores the potential of proteomic profiling to improve cancer treatment by providing targeted, actionable insights into tumor biology.

29 August 2016 in Monatshefte für Chemie - Chemical Monthly
Verified
Journal Article
Closed access
PDF file
Alison Thavary UngAlison Thavary Ung
Alison Thavary Ung
AWAlexander N. West
Alexander N. West
MPMatthew J. A. Phillips
Matthew J. A. Phillips
SWSteven G. Williams
Steven G. Williams