Bayesian tensor factorization-drive breast cancer subtyping by integrating multi-omics data

Liu, Qian; Cheng, Bowen; Jin, Yong Won; Hu, Pingzhao

doi:10.1016/j.jbi.2021.103958

Cited by 17 publications

(5 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Initially, BC was proposed as a single disease originating in the mammary gland. However, it is now established that BC is a complex disease with inter-tumor heterogeneity, and the heterogeneous nature has a significant impact on the progression of the disease and its treatment ( Liu et al, 2022 ). Although the incidence of BC is on the rise all across the world, the mortalities and survival rates vary in different regions, which are attributed to changes in risk factors, hormonal profiles, environmental conditions, access to and standards of healthcare, and genetic features ( Momenimovahed and Salehiniya, 2019 ).…”

Section: Introductionmentioning

confidence: 99%

Clinico-genomic findings, molecular docking, and mutational spectrum in an understudied population with breast cancer patients from KP, Pakistan

Ahmad,

Ali,

Khalil

et al. 2024

Front. Genet.

View full text Add to dashboard Cite

In this study, we report the mutational profiles, pathogenicity, and their association with different clinicopathologic and sociogenetic factors in patients with Pashtun ethnicity for the first time. A total of 19 FFPE blocks of invasive ductal carcinoma (IDC) from the Breast Cancer (BC) tissue and 6 normal FFPE blocks were analyzed by whole-exome sequencing (WES). Various somatic and germline mutations were identified in cancer-related genes, i.e., ATM, CHEK2, PALB2, and XRCC2. Among a total of 18 mutations, 14 mutations were somatic and 4 were germline. The ATM gene exhibited the maximum number of mutations (11/18), followed by CHEK2 (3/18), PALB2 (3/18), and XRCC2 (1/18). Except one frameshift deletion, all other 17 mutations were nonsynonymous single-nucleotide variants (SNVs). SIFT prediction revealed 7/18 (38.8%) mutations as deleterious. PolyPhen-2 and MutationTaster identified 5/18 (27.7%) mutations as probably damaging and 10/18 (55.5%) mutations as disease-causing, respectively. Mutations like PALB2 p.Q559R (6/19; 31.5%), XRCC2 p.R188H (5/19; 26.31%), and ATM p.D1853N (4/19; 21.05%) were recurrent mutations and proposed to have a biomarker potential. The protein network prediction was performed using GeneMANIA and STRING. ISPRED-SEQ indicated three interaction site mutations which were further used for molecular dynamic simulation. An average increase in the radius of gyration was observed in all three mutated proteins revealing their perturbed folding behavior. Obtained SNVs were further correlated with various parameters related to the clinicopathological status of the tumors. Three mutation positions (ATMp. D1853N, CHEK2 p.M314I, and PALB2 p.T1029S) were found to be highly conserved. Finally, the wild- and mutant-type proteins were screened for two drugs: elagolix (DrugBank ID: DB11979) and LTS0102038 (a triterpenoid, isolated from the anticancer medicinal plant Fagonia indica). Comparatively, a higher number of interactions were noted for normal ATM with both compounds, as compared to mutants.

show abstract

Section: Introductionmentioning

confidence: 99%

Clinico-genomic findings, molecular docking, and mutational spectrum in an understudied population with breast cancer patients from KP, Pakistan

Ahmad,

Ali,

Khalil

et al. 2024

Front. Genet.

View full text Add to dashboard Cite

show abstract

“…There is a wide variety of breast cancers. Subtyping the illness and finding the genetic markers that drive these subtypes are essential for precision oncology in breast cancer through study [27]-investigating the possibility of developing a novel computational approach to subtyping breast cancer. The Cancer Genome Atlas assessed 762 breast cancer patients using Bayesian tensor factorization (BTF), a method for integrating multiomics data on breast cancer.…”

Section: Introductionmentioning

confidence: 99%

A Comprehensive Review of Artificial Intelligence Approaches in Omics Data Processing: Evaluating Progress and Challenges

Ali,

Mohammed

2023

ijmscs

View full text Add to dashboard Cite

The primary objective of this study is to review and assess the best available research on omics-related artificial intelligence (AI) methods. Furthermore, it seeks to demonstrate the promise of AI approaches in omics data analysis and identify the critical problems that must be solved to achieve this potential fully. There are many moving parts when trying to make sense of a plethora of research through a literature review. Essential components include, for instance, clinical applications and literature collections. Other researchers have faced challenges, and the existing literature highlights them. Using a systematic strategy, we searched all relevant articles on omics and AI utilizing multiple keyword variations. We also seek additional research, such as guidelines, studies of comparison, and review studies. Challenges with AI, preprocessing, datasets, validation of models, and testbed applications arose when AI was used to analyze omics data. To solve these problems, several pertinent investigations were carried out. Our work offers unique insights into the intersection of omics and AI model fields, setting it apart from prior review articles. We anticipate that practitioners seeking an all-encompassing perspective on using AI in omics data processing would find this study's findings invaluable.

show abstract

“…Nonnegative matrix factorization 3 (NMF) is a powerful tool for feature learning and dimensionality approximate reduction, and it is widely used in gene expression profile data for tumor recognition [4][5] . The objective function of the traditional NMF model is based on the Euclidean distance metric, and its feature learning ability is vulnerable to the influence of noise.…”

Section: Introductionmentioning

confidence: 99%

Supervised nonnegative matrix factorization model with fused correntropy for tumor recognition

Wei

Liu

Yang

2023

International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2023)

View full text Add to dashboard Cite

Nonnegative matrix factorization (NMF) is a feature learning method that can achieve nonlinear dimensional approximate reduction with strong interpretation, and it is widely used in the field of tumor recognition. The objective function of the traditional NMF model is based on the Euclidean distance metric, and the performance of the model is easily affected by the noise. Moreover, traditional NMF is an unsupervised feature learning method that does not use the label information of the data. However, it would cause a waste of information without using label information and cannot learn the discriminative features in the data. Therefore, the supervised nonnegative matrix factorization model with fused correntropy (FCSNMF) is proposed in this paper. The FCSNMF model alleviates the effect of noise in the experimental data by fusing the Euclidean distance metric and the maximum correntropy metric. In addition, the label consistency regularization term is skillfully chosen to utilize the label information of the data to obtain discriminative features. The effectiveness of the FCSNMF model is verified by applying it to a gene expression profile dataset for tumor recognition.

show abstract

Bayesian tensor factorization-drive breast cancer subtyping by integrating multi-omics data

Cited by 17 publications

References 52 publications

Clinico-genomic findings, molecular docking, and mutational spectrum in an understudied population with breast cancer patients from KP, Pakistan

Clinico-genomic findings, molecular docking, and mutational spectrum in an understudied population with breast cancer patients from KP, Pakistan

A Comprehensive Review of Artificial Intelligence Approaches in Omics Data Processing: Evaluating Progress and Challenges

Supervised nonnegative matrix factorization model with fused correntropy for tumor recognition

Contact Info

Product

Resources

About