Molecular Biology and Genetics / Moleküler Biyoloji ve Genetik
Permanent URI for this collectionhttps://hdl.handle.net/11147/9
Browse
3 results
Search Results
Article Citation - WoS: 3Citation - Scopus: 6Improved Cell Segmentation Using Deep Learning in Label-Free Optical Microscopy Images(TÜBİTAK - Türkiye Bilimsel ve Teknolojik Araştırma Kurumu, 2021) Ayanzadeh, Aydın; Yalçın Özuysal, Özden; Pesen Okvur, Devrim; Önal, Sevgi; Töreyin, Behçet Uğur; Ünay, DevrimThe recently popular deep neural networks (DNNs) have a significant effect on the improvement of segmentation accuracy from various perspectives, including robustness and completeness in comparison to conventional methods. We determined that the naive U-Net has some lacks in specific perspectives and there is high potential for further enhancements on the model. Therefore, we employed some modifications in different folds of the U-Net to overcome this problem. Based on the probable opportunity for improvement, we develop a novel architecture by using an alternative feature extractor in the encoder of U-Net and replacing the plain blocks with residual blocks in the decoder. This alteration makes the model superconvergent yielding improved performance results on two challenging optical microscopy image series: a phase-contrast dataset of our own (MDA-MB-231) and a brightfield dataset from a well-known challenge (DSB2018). We utilized the U-Net with pretrained ResNet-18 as the encoder for the segmentation task. Hence, following the modifications, we redesign a novel skip-connection to reduce the semantic gap between the encoder and the decoder. The proposed skip-connection increases the accuracy of the model on both datasets. The proposed segmentation approach results in Jaccard Index values of 85.0% and 89.2% on the DSB2018 and MDA-MB-231 datasets, respectively. The results reveal that our method achieves competitive results compared to the state-of-the-art approaches and surpasses the performance of baseline approaches.Article Citation - WoS: 24Citation - Scopus: 23On-Chip Determination of Tissue-Specific Metastatic Potential of Breast Cancer Cells(Wiley, 2021) Fıratlıgil Yıldırır, Burcu; Batı Ayaz, Gizem; Tahmaz, İsmail; Bilgen, Müge; Pesen Okvur, Devrim; Yalçın Özuysal, ÖzdenMetastasis is one of the major obstacles for breast cancer patients. Limitations of current models demand the development of custom platforms to predict metastatic potential and homing choices of cancer cells. Here, two organ-on-chip platforms, invasion/chemotaxis (IC-chip) and extravasation (EX-chip) were used for the quantitative assessment of invasion and extravasation towards specific tissues. Lung, liver and breast microenvironments were simulated in the chips using tissue-specific cells embedded in matrigel. In the IC-chip, invasive MDA-MB-231, but not noninvasive MCF-7 breast cancer cells invaded into lung and liver microenvironments. In the EX-chip, MDA-MB-231 cells extravasated more into the lung compared to the liver and breast microenvironments. In addition, lung-specific MDA-MB-231 clone invaded and extravasated into the lung microenvironment more efficiently than the bone-specific clone. Both invasion/chemotaxis and extravasation results were in agreement with published clinical data. Collectively, our results show that IC-chip and EX-chip, simulating tissue-specific microenvironments, can distinguish different in vivo metastatic phenotypes, in vitro. Determination of tissue-specific metastatic potential of breast cancer cells is expected to improve diagnosis and help select the ideal therapy.Article Citation - WoS: 1Cellular Distribution of Invadopodia Is Regulated by Nanometer Scale Surface Protein Patterns(Elsevier Ltd., 2017) Batı Ayaz, Gizem; Can, Ali; Pesen Okvur, DevrimInvadopodia are proteolytic structures formed by cancer cells. It is not known whether their cellular distribution can be regulated by the organization of the extracellular matrix or the organization of the golgi complex or whether they have an adhesion requirement. Here, we used electron beam lithography to fabricate fibronectin (FN) nanodots with isotropic and gradient micrometer scale spacings on K-casein and laminin backgrounds. Investigating cancer cells cultured on protein nanopatterns, we showed that (i) presence of FN nanodots on a K-casein background decreased percent of cells with neutral invadopodia polarization compared to FN control surfaces; (ii) presence of a gradient of FN nanodots on a K-casein background increased percent of cells with negative invadopodia polarization compared to FN control surfaces; (iii) polarization of the golgi complex was similar to that of invadopodia in agreement with a spatial link; (iv) local adhesion did not necessarily appear to be a prerequisite for invadopodia formation.