Sharif Digital Repository

Triple Negative Breast Cancer (TNBC) is an invasive subtype of breast cancer. Finding prognostic biomarkers is helpful in choosing the appropriate treatment procedure for patients of this cancer. In recent years, the role of microRNAs in various biological processes, including cancer, has been identified, and their accessibility and stability have made these types of molecules an ideal biomarker. In the first phase of this study, with the aim of overcoming the limitations of previous studies, a new bioinformatics protocol has been proposed to investigate the prognostic miRNAs of triple negative breast cancer. First, using survival analysis, 56 prognostic miRNAs which had a significant... 

Unsupervised neural spike sorting is a crucial tool in studying neural systems in the resolution of a neural cell. In extracellular recording from neural cells, the voltage of media is captured by the electrodes. The situation is possible that an electrode record activity of multiple neurons at the same time. The spike sorting goal is assigning each spike (extracellular recorded neural action potential) to a neural cell that generates it. Conventionally, more than one electrode is used to recording media voltages. The electrodes are placed in a small space as a single device called a multi-electrode array. After the spike sorting procedure, the occurrence time of activity of several cells is... 

Traditional techniques use invasive histology techniques to diagnose cancer. Cancer tissue is sampled directly in this method, which is very painful for the patient. In recent years, scientists have discovered that the cell world is released into the blood plasma after cell death, obtaining useful cancer information. Since methylation changes in cancer cells are very significant and the death rate of cancer cells is high, the methylation of each tissue is different from the other. Furthermore, they were diagnosing the type of cancer.On the other hand, due to the different patterns in methylated DNA with normal DNA and the use of bisulfite treatment technique to detect the degree of... 

The advent and advance of single-cell technologies have enabled us to measure the cell function and identity by using different assays and viewing it by different technologies. Nowadays, we are able to measure multiple feature vectors from same- single cells from multiple abstract molecular levels (genome, transcriptome, proteome, ...) simultaneously. Hence, the analysts can view the cell from different yet correlated angles and study their behaviours. Such progress in joint single-cell assessments plus the development and spread of more common single-cell assays - that measure one feature vector per cell - caused the growing need for computational tools to integrate these datasets in order... 

With the growing use of machine learning algorithms, especially in deep neural networks, the need for annotated data for supervised learning has also increased. In many cases, it is possible to collect data widely, but annotating all of these data is usually very time-consuming, expensive, and even impossible in some cases. The goal of active learning algorithms is to maximize the model’s performance with the least annotated data. Active learning algorithms are iterative algorithms that train the model in each iteration with the current annotated data. Then, using the results of the model on the remaining data without annotation, select some new data to annotate. This process usually... 

The Major Histocompatibility Complex (MHC) Class I molecules play a crucial role in the immune system. These molecules present peptides derived from intracellular proteins on the cell surface to be recognized by T cells. This process is vital for identifying and eliminating cancerous or infected cells. In cancer therapy, particularly in the development of personalized vaccines, accurately selecting peptides that can effectively bind to MHC Class I and stimulate a strong immune response is a significant challenge. This research introduces an innovative neural network model that utilizes a cross-encoder architecture and leverages a pre-trained model to simultaneously process peptide and MHC... 

One of the common needs in bioinformatics research is the detection of shared features and similarities between read sequences or with a reference genome. For this purpose, both precise and heuristic algorithms are available. Generally, detecting similarities among various bioinformatics patterns is crucial for identifying and characterizing biological behaviors. One of the algorithms for estimating similarity between two large sets is locality-sensitive hashing (LSH), which has seen significant application over the past two decades. This algorithm has been utilized in bioinformatics across various areas, including genome assembly, alignment of reads to the reference genome, and metagenomic... 

DNA methylation in one of the most important epigenetic variations, which causes significant variations in gene expressions of mammalians. Our current knowledge about DNA methylation is based on measurments from samples of bulk data which cause ambiguity in intracellular differences and analysis of rare cell samples. For this reason, the ability to measure DNA methylation in single-cells has the potential to play an important role in understanding many biological processes including embryonic developement, disease progression including cancer, aging, chromosome instability, X chromosome inactivation, cell differentiation and genes regulation. Recent technological advances have enabled... 

The human leukocyte antigen protein, commonly known as HLA, has the ability to present small protein fragments called peptides on the surface of cells, whether they originate from within the cell or externally. The binding of these peptides to HLA receptors is a crucial step that triggers an immune response. By estimating the affinity between peptides and HLA class I, we can identify novel antigens that have the potential to be targeted by cancer therapeutic vaccines. Computational methods that predict the binding affinity between peptides and HLA receptors have the potential to expedite the design process of cancer vaccines. Currently, most computational methods exclusively rely on... 

In recent years, monoclonal antibodies have gained attention as highly effective drugs for treating diseases, especially cancer. The high binding affinity between an antibody and its corresponding antigen is one of the key factors in triggering an effective immune response. Modeling binding affinity using machine learning is considered a promising and cost-effective computational approach; however, due to the lack of training data, the performance of these models is often poor and limited. In contrast, recent advances in geometric learning have demonstrated that incorporating the three-dimensional geometry of protein structures in the learning process can significantly impact 3D... 

Genome mutations whose effects are not specified pose one of the challenges in identifying genetic diseases. Utilizing wet lab tests to detect the pathogenicity of variants can be time-consuming and fiscally expensive. A rapid and cost-effective solution to this problem is the use of machine learning-based variant effect predictors, which have the ability to determine whether a mutation is pathogenic or not. The objective of this research is to predict the pathogenicity of genome variations. The proposed model exclusively utilizes the protein sequence as its input feature and does not have access to other protein features. The data used to construct the model comprises mutations with... 

This research addresses the challenge of controllability in computational antibody design using generative diffusion models, particularly RFantibody. Although these models have demonstrated success in generating diverse antibody structures, they still face limitations in steering the generative process toward specific design goals and directly optimizing binding affinity. The main innovation of this study lies in introducing two external guiding potentials during inference, which enhance the generation process in a modular manner without retraining the model. The first potential, focusing on interface hotspots, directs the antibody toward the desired epitope. The second potential, inspired... 

The epidemic of Covid 19 has killed millions of people worldwide. Despite the efforts of scientists around the world, there is still no cure for this disease. Approval of newly designed drugs due to clinical trial periods is time-consuming and costly. For this reason, in the current emergency situation, it is important to have a solution for screening available approved drugs in order to find effective substances for this disease.High-throughput assays are a good option for such problems. In this field of research, image-based high-throughput assays are amongst the most effective and cost-effective methods that help quantify the response of treated cells by measuring cell... 

From a long time ago cancer has been threatening human’s health, and researchers have been grappling with the phenomenon for numerous years. In the annals of this struggle, the number of cancer victims has outnumbered the survivals in a way that,until recently, suffering from cancer was perceived to be equivalent to death. Permanent defeat against cancer stems from the incomplete recognition of the phenomenon. In recent years, with the advent of technologies to extract information from the heart of cells and at the genome and transcriptome levels, man has been able to acquire a deeper understanding of cancer, its behavior and operation. Now that cancer is regarded to be a genetic disease,... 

This study explores the problem of multimodal information retrieval in English and Persian, with a particular focus on electronic commerce where users often search for products using textual or visual queries. To enable multimodal retrieval in Persian, a Persian CLIP model (Contrastive Language-Image Pretraining) was trained. The training dataset was constructed from multiple sources, including machine-translated English corpora, web-crawled Persian data from the Divar platform, and user search and click logs from the Torob marketplace. After filtering and quality evaluation, the model successfully learned a shared embedding space between Persian text and images, forming the foundation for a... 

Histopathology images are a type of medical images that are used to diagnose a variety of diseases. One of these illnesses is the Leukemia cancer, which has four different subtypes and is diagnosed using a blood smear image. As a result of the advancement of deep learning tools, models for diagnosing various types of disease from images have been developed in recent years.In this project, one of the best models developed to diagnose four different types of disease was replicated, and it was demonstrated that, while this model achieves acceptable accuracy, its decision is not based on medically significant criteria. In the following, a general method for diagnosing the disease is proposed...