It is difficult for the parotid gland neoplasms to make an accurate preoperative diagnosis due to the restriction of biopsy in the parotid gland neoplasms. The aim of this study is to apply the surface-enhanced Raman spectroscopy (SERS) method for the blood serum biochemical detection and use the support vector machine for the analysis in order to develop a simple but accurate blood serum detection for preoperative diagnosis of the parotid gland neoplasms. The blood serums were collected from four groups: the patients with pleomorphic adenoma, the patients with Warthin’s tumor, the patients with mucoepidermoid carcinoma and the volunteers without parotid gland neoplasms. Au nanoparticles (Au NPs) were mixed with the blood serum as the SERS active nanosensor to enhance the Raman scattering signals produced by the various biochemical materials and high quality SERS spectrum were obtained by using the Raman microscope system. Then the support vector machine was utilized to analyze the differences of the SERS spectrum from the blood serum of different groups and established a diagnostic model to discriminate the different groups. It was demonstrated that there were different intensities of SERS peaks assigned to various biochemical changes in the blood serum between the parotid gland tumor groups and normal control group. Compared with the SERS spectra of the normal serums, the intensities of peaks assigned to nucleic acids and proteins increased in the SERS spectra of the parotid gland tumor serums, which manifested the differences of the biochemical metabolites in the serum from the patients with parotid gland tumors. When the leave-one-sample-out method was used, support vector machine (SVM) played an outstanding performance in the classification of the SERS spectra with the high accuracy (84.1 % ~ 88.3 %), sensitivity (82.2 % ~ 97.4 %) and specificity (73.7 % ~ 86.7 %). Though the accuracy, sensitivity and specificity decreased in the leave-one-patient-out cross validation, the mucoepidermoid carcinoma was still easier to diagnose than other tumors. The specific molecular differences of parotid gland tumors and normal serums were significantly demonstrated through the comparison between the various SERS spectra.But compared with the serum SERS spectra reported in the other studies, some differences exist between the spectra in this study and the ones reported in the lietratures. These differences may result from the various nano-particles, the different preparation of serum and equipment parameters, and we could need a further research to find an exact explanation.Based on the SERS spectra of the serum samples, SVM have shown a giant potential to diagnose the parotid gland tumors in our preliminary study. However, different cross validaiton methods could effect the accuracy and a further study involing a great number of samples should be needed. This exploratory research demonstrated the great potential of SERS combined with SVM into a non-invasive clinical diagnostic method for preoperative diagnosis of parotid gland tumors. And the internal relation between the spectra and patients should be established in the further study.

Reliability and Reproducibility of differentially expressed genes (DEGs) are essential for the biological interpretation of microarray data. The microarray quality control (MAQC) project launched by US Food and Drug Administration (FDA) elucidated that the lists of DEGs generated by intra- and inter-platform comparisons can reach a high level of concordance, which mainly depended on the statistical criteria used for ranking and selecting DEGs. Generally, it will produce reproducible lists of DEGs when combining fold change ranking with a non-stringent p-value cutoff. For further interpretation of the gene expression data, statistical methods of gene enrichment analysis provide powerful tools for associating the DEGs with prior biological knowledge, e.g. Gene Ontology (GO) terms and pathways, and are widely used in genome-wide research. Although the DEG lists generated from the same compared conditions proved to be reliable, the reproducible enrichment results are still crucial to the discovery of the underlying molecular mechanism differentiating the two conditions. Therefore, it is important to know whether the enrichment results are still reproducible, when using the lists of DEGs generated by different statistic criteria from inter-laboratory and cross-platform comparisons. In our study, we used the MAQC data sets for systematically accessing the intra- and inter-platform concordance of GO terms enriched by Gene Set Enrichment Analysis (GSEA) and LRpath.

The accurate prediction of adverse drug reactions (ADRs) is essential for comprehensive drug safety evaluation. Pre-trained deep chemical language models have emerged as powerful tools capable of automatically learning molecular structural features from large-scale datasets, showing promising capabilities for the downstream prediction of molecular properties. However, the performance of pre-trained chemical language models in predicting ADRs, especially idiosyncratic ADRs induced by marketed drugs, remains largely unexplored. In this study, we propose MoLFormer-XL, a pre-trained model for encoding molecular features from canonical SMILES, in conjunction with a CNN-based model to predict drug-induced QT interval prolongation (DIQT), drug-induced teratogenicity (DIT), and drug-induced rhabdomyolysis (DIR). Our results demonstrate that the proposed model outperforms conventional models applied in previous studies for predicting DIQT, DIT, and DIR. Notably, an analysis of the learned linear attention maps highlights amines, alcohol, ethers, and aromatic halogen compounds as strongly associated with the three types of ADRs. These findings hold promise for enhancing drug discovery pipelines and reducing the drug attrition rate due to safety concerns.

Background: Si-Wu-Tang (SWT), comprising the combination of four herbs, Paeoniae, Angelicae, Chuanxiong and Rehmanniae, is one of the most popular traditional oriental medicines for women’s diseases. In our previous study, the microarray gene expression profiles of SWT on breast cancer cell line MCF-7 were found similar to the effect of β-estradiol (E2) on MCF-7 cells in the Connectivity Map database. Methods: Further data analysis was conducted to find the main similarities and differences between the effects of SWT and E2 on MCF-7 gene expression. The cell proliferation assay on MCF-7 (ER-positive) and MDA-MB-231 (ER-negative) cells were used to examine such estrogenic activity. The estrogenic potency of SWT was further confirmed by estrogen-responsive element (ERE) luciferase reporter assay in MCF-7 cells. Results: Many estrogen regulated genes strongly up-regulated by E2 were similarly up-regulated by SWT, e.g., GREB1, PGR and EGR3. Of interest with regard to safety of SWT, the oncogenes MYBL1 and RET were strongly induced by E2 but not by SWT. Quantitative RT-PCR analysis revealed a highly concordant expression change in selected genes with data obtained by microarrays. Further supporting SWT’s estrogenic activity, in MCF-7 but not in MDA-MB-231 cells, SWT stimulated cell growth at lower concentrations (< 3.0 mg/ml), while at high concentrations, it inhibits the growth of both cell lines. The growth inhibitory potency of SWT was significantly higher in MDA-MB-231 than in MCF-7 cells. The SWT-induced cell growth of MCF-7 could be blocked by addition of the estrogen receptor antagonist tamoxifen. In addition, SWT was able to activate the ERE activity at lower concentrations. The herbal components Angelicae, Chuanxiong and Rehmanniae at lower concentrations (< 3.0 mg/ml) also showed growth-inducing and ERE-activating activity in MCF-7 cells. Conclusions: These results revealed a new mechanism to support the clinical use of SWT for estrogen related diseases and possibly for cancer prevention. This study also demonstrated the feasibility of using microarray transcriptional profiling to discover phytoestrogenic components that are present in natural products.