Protein subcellular localization based on deep image features and criterion learning strategy.

The spatial distribution of proteome at subcellular levels provides clues for protein functions, thus is important to human biology and medicine. Imaging-based methods are one of the most important approaches for predicting protein subcellular location. Although deep neural networks have shown impressive performance in a number of imaging tasks, its application to protein subcellular localization has not been sufficiently explored. In this study, we developed a deep imaging-based approach to localize the proteins at subcellular levels. Based on deep image features extracted from convolutional neural networks (CNNs), both single-label and multi-label locations can be accurately predicted. Particularly, the multi-label prediction is quite a challenging task. Here we developed a criterion learning strategy to exploit the label-attribute relevancy and label-label relevancy. A criterion that was used to determine the final label set was automatically obtained during the learning procedure. We concluded an optimal CNN architecture that could give the best results. Besides, experiments show that compared with the hand-crafted features, the deep features present more accurate prediction with less features. The implementation for the proposed method is available at https://github.com/RanSuLab/ProteinSubcellularLocation.

[Recognition of corn seeds based on pattern recognition and near infrared spectroscopy technology].

Pattern recognition technology and data mining methods have become a hot topic in chemometrics. Near infrared (NIR) spectroscopic analysis has been widely used in spectrum signal processing and modeling due to its advantages of quickness, simplicity and nondestructiveness. Based on five different methods of pattern recognition, namely the locally linear embedding (LLE), wavelet transform (WT), principal component analysis (PCA), partial least squares (PLS) and support vector machine (SVM), the pattern recognition system for corn seeds is proposed using NIR technology, and applied to classification of 108 hybrid samples and 178 female samples for corn seeds. Firstly, we get rid of noise or reduce the dimension using LLE, WT, PCA and PLS, and then use SVM to identify two-class samples. In the meantime, 1-norm SVM is the method of direct classification and identification. Experimental results for three different spectral regions show that the performances of three methods, i. e. PCA+SVM, LLE+SVM, PLS+SVM, are superior to WT+SVM and 1-norm SVM methods, and obtain a high classification accuracy, which indicates the feasibility and effectiveness of the proposed methods. Moreover, this investigation provides the theoretical support and practical method for recognition of corn seeds utilizing near infrared spectral data.

[Recognition of corn seeds based on pattern recognition and near infrared spectroscopy technology].

Pattern recognition technology and data mining methods have become a hot topic in chemometrics. Near infrared (NIR) spectroscopic analysis has been widely used in spectrum signal processing and modeling since it has advantages of quickness, simplicity and nondestructiveness. Based on five different methods of pattern recognition, namely the locally linear embedding (LLE), wavelet transform (WT), principal component analysis (PCA), partial least squares (PLS) and support vector machine (SVM), the pattern recognition system for corn seeds was proposed using NIR technology, and applied to classification of 108 hybrid samples and 178 female samples for corn seeds. Firstly, we get rid of noise or reduce the dimension using LLE, WT, PCA, PLS, and then use SVM to identify two-class samples. In the meantime, 1-norm SVM is the method of direct classification and identification. Experimental results of three different spectral regions show that the performances of three methods: PCA+SVM, LLE+SVM, PLS+SVM are superior to WT+SVM and 1-norm SVM methods, and obtain a high classification accuracy, which indicates the feasibility and effectiveness of the proposed methods. Moreover, this investigation provides the theoretical support and practical method for recognition of corn seeds utilizing near infrared spectral data.

Identification of a novel SEREX antigen, SLC2A1/GLUT1, in esophageal squamous cell carcinoma.

We have carried out SEREX (serological identification of antigens by recombinant cDNA expression cloning), and identified SLC2A1 (solute carrier family 2/facilitated glucose transporter, member 1) as an antigen recognized by serum IgG antibodies in patients with esophageal squamous cell carcinoma (SCC). The levels of serum anti-SLC2A1 antibodies (s-SLC2A1-Abs), examined by enzyme-linked immunosorbent assay using bacterially expressed glutathione-S-transferase-SLC2A1 fusion protein, were significantly higher in patients with esophageal SCC than in healthy donors. When using a cut-off level as the mean + 2x standard deviations of healthy donors, a total of 12 (21%) out of 57 SCC patients were revealed as positive for s-SLC2A1-Abs. The presence of s-SLC2A1-Abs was not associated with either clinicopathological factors or survival. Because s-SLC2A1-Abs were not associated with the positivity of other conventional serum markers, a combination assay of s-SLC2A1-Abs with these conventional serum markers may be useful for the diagnosis and monitoring of esophageal SCC.

Facilitation of adenoviral wild-type p53-induced apoptotic cell death by overexpression of p33(ING1) in T.Tn human esophageal carcinoma cells.

To investigate the effect of p33(ING1) on wild-type p53 gene therapy, T.Tn human esophageal carcinoma cells were stably transfected with p33(ING1) cDNA. Infection with Ad-p53 (recombinant adenovirus containing wild-type p53) into p33-transfected cells reduced cell viability, while infection with empty vector had little effect. This reduced viability was shown to be due to apoptotic cell death by the TUNEL (terminal deoxynucleotidyl transferase-mediated nick end-labeling) assay. Following infection with Ad-p53, levels of p53 were similar in p33-expressing cells and in the parental line. However, levels of p21 and Mdm2 were elevated in p33-transfected cells. Nonetheless, this enhanced expression of Mdm2 appeared to be ineffective in downregulating p53. Transient transfection with mutant Mdm2 prior to Ad-p53 infection provided a significant protection as compared with cells transfected with wild-type Mdm2. These results imply a synergistic effect between p33 and p53 in the induction of apoptosis of human esophageal carcinoma cells. A role for Mdm2 in this synergism is suggested.

Decrease in growth factor receptors after treatment with serine protease inhibitor ONO-3403.

FOY-305 is a synthetic serine protease inhibitor and ONO-3403 is a derivative with a higher protease-inhibitory activity. The growth-suppressive effects of ONO-3403 were more prominent in Ha-ras-transformed NIH3T3 (ras-NIH) cells than in non-transformed NIH3T3 cells. After treatment of ras-NIH cells with ONO-3403 at 100-200 microg/ml, the percentage of cells found in G(1) phase decreased and, concomitantly, that in S phase increased. Molecular events caused by ONO-3403 were investigated by Western blot analysis using anti-phosphotyrosine antibody. The results showed a marked decrease in the tyrosine phosphorylation level of a 180-kDa protein after treatment with ONO-3403. This 180-kDa phosphotyrosine-containing molecule which was tentatively designated pY-p180 might be platelet-derived growth factor (PDGF) receptor since addition of PDGF to serum-starved NIH3T3 cells induced a marked tyrosine phosphorylation of the same size within 5 min. This was further confirmed by immunoprecipitation of cell extract with anti-PDGF-receptor antibody followed by Western blot analysis using anti-phosphotyrosine antibody. Treatment of T.Tn human esophageal carcinoma cells with ONO-3403 caused also decrease in pY-p180, which appeared to be epidermal growth factor receptor. Thus, ONO-3403 may induce growth suppression by down-regulation of cell surface growth factor receptors.

Presence of serum tripartite motif-containing 21 antibodies in patients with esophageal squamous cell carcinoma.

SEREX has been applied to esophageal SCC, and the TRIM21 gene was identified as a novel SEREX antigen of esophageal SCC. The presence of s-TRIM21-Abs was confirmed by Western blotting using bacterially expressed TRIM21 gene product and was evaluated for clinicopathological significance in patients with esophageal SCC. s-TRIM21-Abs were detected in 18 (20%) of 91 patients with esophageal SCC but not in 42 healthy donors. The presence of s-TRIM21-Abs was partly associated with tumor size (P = 0.063) and poor survival (P = 0.067). To measure serum antibody levels, ELISA using purified recombinant TRIM21 protein was developed. The levels of s-TRIM21-Abs were significantly higher in patients with esophageal SCC than in healthy donors (P = 0.013). s-TRIM21-Abs may be a useful tumor marker to diagnose and predict disease progression in patients with esophageal SCC.

Increased expression of collagen XVIII and its prognostic value in nonsmall cell lung carcinoma.

BACKGROUND: Angiogenesis plays a crucial role in tumor growth and metastasis. Recently, some studies have focused on the angiogenesis inhibitor endostatin. However, the biologic role of the precursor of endostatin, collagen XVIII, in human malignancy is unknown. The purpose of the current study was to evaluate whether the expression of collagen XVIII has additional prognostic value for survival in patients with nonsmall cell lung carcinoma (NSCLC). METHODS: The authors investigated the expression of collagen XVIII in 221 patients using immunohistochemical methods. To confirm the specificity of the collagen XVIII polyclonal antibody used in the current study and to test the expression of collagen XVIII in human lung carcinoma, Western blot analysis was performed on a panel of human lung carcinoma cell lines. RESULTS: Collagen XVIII expression was detected in 162 of 221 patients with NSCLC (73%), primarily in the tumor cell cytoplasm. Low collagen XVIII expression levels were found in 75 tumor specimens, while high collagen XVIII expression levels were noted in 87 tumor specimens. The prevalence of positive collagen XVIII expression was greater in T2-4 tumors than in T1 tumors (P = 0.0235). The prognosis for patients with strongly collagen XVIII-positive NSCLC was significantly worse than the prognosis for patients with collagen XVIII-positive or collagen XVIII-negative NSCLC (P = 0.0010). Multivariate analysis indicated that T status, lymph node status, and the overexpression of collagen XVIII were independent prognostic factors. CONCLUSIONS: The results of the current study indicated that the overexpression of collagen XVIII was associated with NSCLC progression and poor outcome. Thus, collagen XVIII expression may serve as a useful prognostic marker in patients with NSCLC.