Differential translation of mRNA isoforms underlies oncogenic activation of cell cycle kinase Aurora A.

Aurora Kinase A (AURKA) is an oncogenic kinase with major roles in mitosis, but also exerts cell cycle- and kinase-independent functions linked to cancer. Therefore, control of its expression, as well as its activity, is crucial. A short and a long 3'UTR isoform exist for AURKA mRNA, resulting from alternative polyadenylation (APA). We initially observed that in triple-negative breast cancer, where AURKA is typically overexpressed, the short isoform is predominant and this correlates with faster relapse times of patients. The short isoform is characterized by higher translational efficiency since translation and decay rate of the long isoform are targeted by hsa-let-7a tumor-suppressor miRNA. Additionally, hsa-let-7a regulates the cell cycle periodicity of translation of the long isoform, whereas the short isoform is translated highly and constantly throughout interphase. Finally, disrupted production of the long isoform led to an increase in proliferation and migration rates of cells. In summary, we uncovered a new mechanism dependent on the cooperation between APA and miRNA targeting likely to be a route of oncogenic activation of human AURKA.

Prediction of Pentacam image after corneal cross-linking by linear interpolation technique and U-NET based 2D regression model.

Keratoconus is a common corneal disease that causes vision loss. In order to prevent the progression of the disease, the corneal cross-linking (CXL) treatment is applied. The follow-up of keratoconus after treatment is essential to predict the course of the disease and possible changes in the treatment. In this paper, a deep learning-based 2D regression method is proposed to predict the postoperative Pentacam map images of CXL-treated patients. New images are obtained by the linear interpolation augmentation method from the Pentacam images obtained before and after the CXL treatment. Augmented images and preoperative Pentacam images are given as input to U-Net-based 2D regression architecture. The output of the regression layer, the last layer of the U-Net architecture, provides a predicted Pentacam image of the later stage of the disease. The similarity of the predicted image in the final layer output to the Pentacam image in the postoperative period is evaluated by image similarity algorithms. As a result of the evaluation, the mean SSIM (The structural similarity index measure), PSNR (peak signal-to-noise ratio), and RMSE (root mean square error) similarity values are calculated as 0.8266, 65.85, and 0.134, respectively. These results show that our method successfully predicts the postoperative images of patients treated with CXL.

Scat-NET: COVID-19 diagnosis with a CNN model using scattergram images.

The acute respiratory syndrome COVID-19 disease, which is caused by SARS-CoV-2, has infected many people over a short time and caused the death of more than 2 million people. The gold standard in detecting COVID-19 is to apply the reverse transcription polymerase chain reaction (RT-PCR) test. This test has low sensitivity and produces false results of approximately 15%-20%. Computer tomography (CT) images were checked as a result of suspicious RT-PCR tests. If the virus is not infected in the lung, the virus is not observed on CT lung images. To overcome this problem, we propose a 25-depth convolutional neural network (CNN) model that uses scattergram images, which we call Scat-NET. Scattergram images are frequently used to reveal the numbers of neutrophils, eosinophils, basophils, lymphocytes and monocytes, which are measurements used in evaluating disease symptoms, and the relationships between them. To the best of our knowledge, using the CNN together with scattergram images in the detection of COVID-19 is the first study on this subject. Scattergram images obtained from 335 patients in total were classified using the Scat-NET architecture. The overall accuracy was 92.4%. The most striking finding in the results obtained was that COVID-19 patients with negative RT-PCR tests but positive CT test results were positive. As a result, we emphasize that the Scat-NET model will be an alternative to CT scans and could be applied as a secondary test for patients with negative RT-PCR tests.

Discrimination of ß-thalassemia and iron deficiency anemia through extreme learning machine and regularized extreme learning machine based decision support system.

The symptoms of Iron Deficiency Anemia (IDA) and ß-thalassemia (ß-TT) disease are similar and the distinction between them is time consuming and costly. There are several indices used to differentiate IDA from ß-thalassemia disease. Complete Blood Count (CBC) is a rapid, inexpensive and accessible test for the diagnosis of anemia and is used as a primary test. However, since CBC cannot fully distinguish between IDA and ß-thalassemia, more advanced testing is required. These tests are not available in small centers and are performed on higher-cost devices. Moreover, it is important to differentiate between anemia and ß-thalassemia medically for two reasons (IDA). First, if a patient with ß-Thalassemia is diagnosed with IDA, the patient is given unnecessary iron supplementation as a result of the treatment, which is recommended by the doctor. Secondly, when the patient with ß-thalassemia is diagnosed with IDA, children will have ß-thalassemia patients in marriages. A decision support system to distinguish between ß-Thalassemia and IDA has been developed. Logistic Regression, K-Nearest Neighbours, Support Vector Machine, Extreme Learning Machine and Regularized Extreme Learning Machine classification algorithms were used in the proposed system. Classification performance was evaluated with Accuracy, sensitivity, f-measure, Specificty parameters using Hemoglobin, RBC, HCT, MCV, MCH, MCHC and RDW parameters obtained from 342 patients. 96.30% accuracy for female, 94.37% for male, and 95.59% in co-evaluation of male and female patients were obtained.

3'UTR shortening and EGF signaling: implications for breast cancer.

Alternative polyadenylation (APA) plays a role in gene expression regulation generally by shortening of 3'UTRs (untranslated regions) upon proliferative signals and relieving microRNA-mediated repression. Owing to high proliferative indices of triple negative breast cancers (TNBCs), we hypothesized APA to cause 3'UTR length changes in this aggressive subgroup of breast cancers. Our probe-based meta-analysis approach identified 3'UTR length alterations where the significant majority was shortening events (∼70%, 113 of 165) of mostly proliferation-related transcripts in 520 TNBC patients compared with controls. Representative shortening events were further investigated for their microRNA binding potentials by computational predictions and dual-luciferase assay. In silico-predicted 3'UTR shortening events were experimentally confirmed in patient and cell line samples. To begin addressing the underlying mechanisms, we found CSTF2 (cleavage stimulation factor 2), a major regulator of 3'UTR shortening to be up-regulated in response to epidermal growth factor (EGF). EGF treatment also resulted with further shortening of the 3'UTRs. To investigate the contribution of CSTF2 and 3'UTR length alterations to the proliferative phenotype, we showed pharmacological inhibition of the EGF pathway to lead to a reduction in CSTF2 levels. Accordingly, RNAi-induced silencing of CSTF2 decreased the proliferative rate of cancer cells. Therefore, our computational and experimental approach revealed a pattern of 3'UTR length changes in TNBC patients and a potential link between APA and EGF signaling. Overall, detection of 3'UTR length alterations of various genes may help the discovery of new cancer-related genes, which may have been overlooked in conventional microarray gene expression analyses.