Metabolomics patterns of breast cancer tumors using mass spectrometry imaging

Theriault, R.; Kaufmann, Martin; Ren, Kevin; Varma, Sonal; Ellis, Randy E.

doi:10.1007/s11548-021-02387-0

Cited by 10 publications

(5 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Deep learning techniques as well as voting mechanisms can also be used to enrich data results [ 24 , 25 ]. E.g., voting mechanisms have been used in [ 16 , 17 ] in order to improve proteins identification results in different contexts. Thus, using machine learning pipelines is not new in the field.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Machine learning pipeline to analyze clinical and proteomics data: experiences on a prostate cancer case

Vizza,

Aracri,

Guzzi

et al. 2024

BMC Med Inform Decis Mak

View full text Add to dashboard Cite

Proteomic-based analysis is used to identify biomarkers in blood samples and tissues. Data produced by devices such as mass spectrometry requires platforms to identify and quantify proteins (or peptides). Clinical information can be related to mass spectrometry data to identify diseases at an early stage. Machine learning techniques can be used to support physicians and biologists in studying and classifying pathologies. We present the application of machine learning techniques to define a pipeline aimed at studying and classifying proteomics data enriched using clinical information. The pipeline allows users to relate established blood biomarkers with clinical parameters and proteomics data. The proposed pipeline entails three main phases: (i) feature selection, (ii) models training, and (iii) models ensembling. We report the experience of applying such a pipeline to prostate-related diseases. Models have been trained on several biological datasets. We report experimental results about two datasets that result from the integration of clinical and mass spectrometry-based data in the contexts of serum and urine analysis. The pipeline receives input data from blood analytes, tissue samples, proteomic analysis, and urine biomarkers. It then trains different models for feature selection, classification and voting. The presented pipeline has been applied on two datasets obtained in a 2 years research project which aimed to extract hidden information from mass spectrometry, serum, and urine samples from hundreds of patients. We report results on analyzing prostate datasets serum with 143 samples, including 79 PCa and 84 BPH patients, and an urine dataset with 121 samples, including 67 PCa and 54 BPH patients. As results pipeline allowed to identify interesting peptides in the two datasets, 6 for the first one and 2 for the second one. The best model for both serum (AUC=0.87, Accuracy=0.83, F1=0.81, Sensitivity=0.84, Specificity=0.81) and urine (AUC=0.88, Accuracy=0.83, F1=0.83, Sensitivity=0.85, Specificity=0.80) datasets showed good predictive performances. We made the pipeline code available on GitHub and we are confident that it will be successfully adopted in similar clinical setups.

show abstract

Section: Discussionmentioning

confidence: 99%

“…In [ 12 ] and in [ 13 – 15 ] ML techniques are used to select biomarkers from prostate and ovarian cancer datasets respectively. Voting has been used in [ 16 , 17 ] to improve protein identification in different use cases.…”

Section: Introductionmentioning

confidence: 99%

Machine learning pipeline to analyze clinical and proteomics data: experiences on a prostate cancer case

Vizza,

Aracri,

Guzzi

et al. 2024

BMC Med Inform Decis Mak

View full text Add to dashboard Cite

show abstract

“…Deep learning techniques as well as voting mechanisms can also be used to enrich data results [24,25]. E.g., voting mechanisms have been used in [26,27] in order to improve proteins identification results in different contexts. Thus, using machine learning pipelines is not new in the field.…”

Section: Discussionmentioning

confidence: 99%

Machine Learning Pipeline to analyze Clinical and Proteomics Data: experiences on a prostate cancer case

Tradigo,

Vizza,

Aracri

et al. 2023

Preprint

View full text Add to dashboard Cite

Proteomic-based analysis is used to identify biomarkers in blood samples and tissues. Data produced by devices such as mass spectrometry, requires platforms to identify and quantify proteins (or peptides). Clinical information can be related with mass spectrometry data to identify diseases at an early stage. Machine learning techniques can be used to support physicians and biologists for studying and classifying pathologies. We present the application of machine learning techniques to define a pipeline aimed to study and classify protemics data enriched by means of clinical information. The pipeline allows users to relate established blood biomarkers with clinical parameters and proteomics data. The proposed pipeline entails three main phases: (i) feature selection, (ii) models training and (iii) models ensembling. We report the experience of applying such a pipeline on prostate related diseases. Models have been trained on a datasets which results from the integration of clinical and mass spectrometry based data. The pipeline receives as input data from blood analytes, tissue samples, proteomic analysis and urine biomarkers. It then learns different models for feature selection and classification.The presented pipeline has been applied on two datasets onbtained a 2 years research project which aimed to extract hidden information from mass spectrometry, serum and urine samples from hundreds of patients. We report results on analyzing prostate datasets with 163 samples, including 79 patient, and a urine dataset with 121 patients. As results pipeline allowed to identify interesting peptides in the two datasets, 6 for the first one and 2 for the second one. The best model for both prostate (AUC=0.815, F1=0.8, Specificity=0.75, Sensitivity=0.88) and urine (AUC=0.810, F1=0.824, Specificity=0.805, Sensitivity=0.814) datasets showed good predictive performances.We are confident that the pipeline can be successfully adopted in similar clinical setups.

show abstract

“…They were particularly effective at finding cancer regions that were discontinuously distributed within benign tissue. The authors concluded that, used intraoperatively, the method can acquire signals that distinguish malignant from benign breast tissue in surgically excised tumors and can be suitable for real-time surgical decisions based on cancer margins . Sharaf (2022) used a combined proteomic and untargeted quantitative metabolomic approach for investigating the effects of tamoxifen and trastuzumab therapy (alone or in combination) and found 77 proteins and 85 metabolite abundances to vary with these treatments.…”

Section: Studies Characterizing Metabolomementioning

confidence: 99%

Recent Contributions of Mass Spectrometry-Based “Omics” in the Studies of Breast Cancer

Banerjee,

Hatimuria,

Sarkar

et al. 2023

Chem. Res. Toxicol.

View full text Add to dashboard Cite

Breast cancer (BC) is one of the most heterogeneous groups of cancer. As every biotype of BC is unique and presents a particular "omic" signature, they are increasingly characterized nowadays with novel mass spectrometry (MS) strategies. BC therapeutic approaches are primarily based on the two features of human epidermal growth factor receptor 2 (HER2) and estrogen receptor (ER) positivity. Various strategic MS implementations are reported in studies of BC also involving data independent acquisitions (DIAs) of MS which report novel differential proteomic, lipidomic, proteogenomic, phosphoproteomic, and metabolomic characterizations associated with the disease and its therapeutics. Recently many "omic" studies have aimed to identify distinct subsidiary biotypes for diagnosis, prognosis, and targets of treatment. Along with these, drug-induced-resistance phenotypes are characterized by "omic" changes. These identifying aspects of the disease may influence treatment outcomes in the near future. Drug quantifications and characterizations are also done regularly and have implications in therapeutic monitoring and in drug efficacy assessments. We report these studies, mentioning their implications toward the understanding of BC. We briefly provide the MS instrumentation principles that are adopted in such studies as an overview with a brief outlook on DIA-MS strategies. In all of these, we have chosen a model cancer for its revelations through MS-based "omics".

show abstract

Metabolomics patterns of breast cancer tumors using mass spectrometry imaging

Cited by 10 publications

References 29 publications

Machine learning pipeline to analyze clinical and proteomics data: experiences on a prostate cancer case

Machine learning pipeline to analyze clinical and proteomics data: experiences on a prostate cancer case

Machine Learning Pipeline to analyze Clinical and Proteomics Data: experiences on a prostate cancer case

Recent Contributions of Mass Spectrometry-Based “Omics” in the Studies of Breast Cancer

Contact Info

Product

Resources

About