Machine learning reveals salivary glycopatterns as potential biomarkers for the diagnosis and prognosis of papillary thyroid cancer

Ren, Xiaolin; Shu, Jiwu; Wang, Junhong; Guo, Yi; Zhang, Ying; Yue, Lixin; Yu, Hanjie; Chen, Wentian; Zhang, Chen; Li, Zheng

doi:10.1016/j.ijbiomac.2022.05.194

Cited by 13 publications

(5 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The development of artificial intelligence as machine learning has shown prospects in the evaluation of disease-based diagnostic efficacy ( 53 , 54 ). We follow the idea of differentiating clinical disease by identifying biomarkers through models that look for potential markers associated with CRPA, an approach that has shown credible accuracy and specificity when evaluating diagnostic efficacy through model construction ( 55 ).…”

Section: Discussionmentioning

confidence: 99%

The glycopatterns of Pseudomonas aeruginosa as a potential biomarker for its carbapenem resistance

Dang,

Shu,

Wang

et al. 2023

Microbiol Spectr

Self Cite

View full text Add to dashboard Cite

Multidrug-resistant (MDR) Pseudomonas aeruginosa infections pose a significant challenge to effective treatment. Although carbapenems were once considered the primary therapeutic option for MDR P. aeruginosa , the clinical use of these antibiotics has become increasingly limited, and the exploration of alternative antimicrobial strategies remains necessary. Bacterial surface glycans are critical in response to antibiotics and represent an attractive therapeutic target. However, the understanding of the role of glycan structures in bacterial resistance is currently quite limited. In this study, we used lectin microarrays to analyze the differences in glycan alterations between 53 drug-sensitive P. aeruginosa (DSPA) strains and 57 carbapenem-resistant P. aeruginosa (CRPA) strains obtained from clinical isolates, with the goal of identifying important glycopatterns associated with carbapenem resistance. The results revealed significant differences in the expression levels of glycan structures [e.g., Fucα1–6GlcNAc, α- d -Man, and Fucα1–3(Galβ1–4)GlcNAc] recognized by 20 lectins (e.g., LCA, PSA, and AAL) on the bacterial surface. Furthermore, we applied K -fold cross-validation to determine the optimal parameters and constructed the DSPA and CRPA models using gradient boosting decision tree (GBDT) algorithm. The GBDT model presented the best performance in an independent test cohort, demonstrating that the screened glycopatterns could serve as potential biomarkers for differentiating bacterial resistance. Our study is the first to apply lectin microarrays to monitor carbapenem resistance in clinical P. aeruginosa isolates, and we hope that the methodology used in this work will serve as an important tool for exploring the association of bacterial resistance with glycosylation mechanisms. IMPORTANCE Bacterial surface glycans are an attractive therapeutic target in response to antibiotics; however, current knowledge of the corresponding mechanisms is rather limited. Antimicrobial susceptibility testing, genome sequencing, and MALDI-TOF MS, commonly used in recent years to analyze bacterial resistance, are unable to rapidly and efficiently establish associations between glycans and resistance. The discovery of new antimicrobial strategies still requires the introduction of promising analytical methods. In this study, we applied lectin microarray technology and a machine-learning model to screen for important glycan structures associated with carbapenem-resistant P. aeruginosa . This work highlights that specific glycopatterns can be important biomarkers associated with bacterial antibiotic resistance, which promises to provide a rapid entry point for exploring new resistance mechanisms in pathogens.

show abstract

Section: Discussionmentioning

confidence: 99%

The glycopatterns of Pseudomonas aeruginosa as a potential biomarker for its carbapenem resistance

Dang,

Shu,

Wang

et al. 2023

Microbiol Spectr

Self Cite

View full text Add to dashboard Cite

show abstract

“…The same steps were also applied to the negative allantoic fluid. All the above collections were used for further SDS-PAGE analysis [ 20 , 21 ].…”

Section: Methodsmentioning

confidence: 99%

N-Glycan Profiles of Neuraminidase from Avian Influenza Viruses

Chen,

Ma,

Liu

et al. 2024

Viruses

Self Cite

View full text Add to dashboard Cite

The cleavage of sialic acids by neuraminidase (NA) facilitates the spread of influenza A virus (IV) descendants. Understanding the enzymatic activity of NA aids research into the transmission of IVs. An effective method for purifying NA was developed using p-aminophenyloxamic acid-modified functionalized hydroxylated magnetic particles (AAMPs), and from 0.299 to 0.401 mg of NA from eight IV strains was isolated by 1 mg AAMP. A combination of lectin microarrays and MALDI-TOF/TOF-MS was employed to investigate the N-glycans of isolated NAs. We found that more than 20 N-glycans were identified, and 16 glycan peaks were identical in the strains derived from chicken embryo cultivation. Multi-antennae, bisected, or core-fucosylated N-glycans are common in all the NAs. The terminal residues of N-glycans are predominantly composed of galactose and N-acetylglucosamine residues. Meanwhile, sialic acid residue was uncommon in these N-glycans. Further computational docking analysis predicted the interaction mechanism between NA and p-aminophenyloxamic acid.

show abstract

“…Furthermore, the integration of machine learning and artificial intelligence (AI) in cancer biomarker screening and diagnosis has transformative potential. , Machine learning algorithms can analyze vast amounts of data, recognize hidden patterns, and develop predictive models. They enhance the accuracy of cancer biomarker screening by processing genomic and proteomic data, identifying novel biomarkers, and uncovering subtle patterns.…”

Section: The Importance Of Biomarkers and Their Challenges For Cancer...mentioning

confidence: 99%

3D Printing for Cancer Diagnosis: What Unique Advantages Are Gained?

Zhang

2023

ACS Mater. Au

View full text Add to dashboard Cite

Cancer is a complex disease with global significance, necessitating continuous advancements in diagnostics and treatment. 3D printing technology has emerged as a revolutionary tool in cancer diagnostics, offering immense potential in detection and monitoring. Traditional diagnostic methods have limitations in providing molecular and genetic tumor information that is crucial for personalized treatment decisions. Biomarkers have become invaluable in cancer diagnostics, but their detection often requires specialized facilities and resources. 3D printing technology enables the fabrication of customized sensor arrays, enhancing the detection of multiple biomarkers specific to different types of cancer. These 3D-printed arrays offer improved sensitivity, allowing the detection of low levels of biomarkers, even in complex samples. Moreover, their specificity can be fine-tuned, reducing false-positive and false-negative results. The streamlined and cost-effective fabrication process of 3D printing makes these sensor arrays accessible, potentially improving cancer diagnostics on a global scale. By harnessing 3D printing, researchers and clinicians can enhance early detection, monitor treatment response, and improve patient outcomes. The integration of 3D printing in cancer diagnostics holds significant promise for the future of personalized cancer care.

show abstract

Machine learning reveals salivary glycopatterns as potential biomarkers for the diagnosis and prognosis of papillary thyroid cancer

Cited by 13 publications

References 34 publications

The glycopatterns of Pseudomonas aeruginosa as a potential biomarker for its carbapenem resistance

The glycopatterns of Pseudomonas aeruginosa as a potential biomarker for its carbapenem resistance

N-Glycan Profiles of Neuraminidase from Avian Influenza Viruses

3D Printing for Cancer Diagnosis: What Unique Advantages Are Gained?

Contact Info

Product

Resources

About