Sharif Digital Repository

Diabetes mellitus is a group of disorders where the level of blood sugar remains high for a long period of time. This increase may be due to either reduced insulin secretion from the pancreatic gland, or insulin resistance, or both. Another key reason is the destruction of beta cells due to functional defect in the body’s immune system. Current treatments include controlling diet, insulin injection and pancreatic transplantation, all of which are temporary. For this reason, finding genetic factors participating in the progression of the disease and adapting treatments to these factors are under intensive studies.In this thesis, available information resources including genomic, biological... 

Distributed learning has been a popular area of machine learning for researchers according to having access to an unprecedented amount of data distributed over many clients in a network. Moreover, high dimensional learning, when the dimensions of data are high but the effective features are low, has great usage in learning especially in medical science. In this paper, we attempted to develop a method for learning high-dimensional sparse problems in a distributed manner, whit a star shape network. Our main focus in this research is on optimizing the communication flow in the network. The prominent idea of our proposed method is to extract the main feature in the first step, then continuing... 

Human is diploid specie that inherits a set of chromosomes from their mother and a set from their father. The process of separating the nucleotide content of a set of extracted maternal and paternal chromosomes for an individual or a population is called phasing the genome of the individual or the population. The placement of any two variants relative to each other in diploid species is possible in two forms: cis (placement of both variants on one chromosome), and trans (placement of variants on different chromosomes). Each of these conditions leads to different phenotypes. Thus, understanding how variants are placed relative to each other is a crucial problem in human biology which is... 

Population genetics studies the distribution and changes in allele frequencies under the influence of five main evolutionary processes: natural selection, genetic drift, mutation, gene flow, and recombination. Among these, the recombination process can influence a wide range of biological processes by rearranging genes, repairing DNA structure, and participating actively in cell division mechanisms. Recombination has the ability to create genetic diversity through gene rearrangement, which is the main reason for creating diversity and evolution in organisms. Models such as Hill-Robertson have proven the influential role of recombination in accelerating evolutionary mechanisms. Also,... 

In recent years, with the advancement of genetics technologies, many data from this field have been made available to researchers. Therefore, many analytical problems have been defined for these data. Genome-wide association studies, or GWAS for short, is one of these issues that deals with finding genetic positions affecting traits or diseases. Common approaches to this problem either examine genetic variants one by one or fail to consider the specific structure of genetic data. Also, both mentioned approaches do not provide a guarantee to control the rate of false positives. In this thesis, an attempt has been made to propose a method to solve the GWAS problem by using the new statistical... 

A random vector with binary components that are independent of each other is referred to as a Bernoulli random vector. A Bernoulli Mixture Model (BMM) is a combination of a finite number of Bernoulli models, where each sample is generated randomly according to one of these models. The important challenge is to estimate the parameters of a Bernoulli Mixture Model or to cluster samples based on their source models. This problem has applications in bioinformatics, image recognition, text classification, social networks, and more. For example, in bioinformatics, it pertains to clustering ethnic groups based on genetic data. Many studies have introduced algorithms for solving this problem without... 

One of the complex problems is routing. This problem becomes more difficult and important in certain situations, which cannot be solved straightforwardly. In this thesis, a model for vehicle (ambulance) routing problem during hospital evacuation in disaster conditions is described and solved using Gray Wolf Optimization (GWO) algorithm in combination with a local search algorithm called Great Deluge Algorithm (GDA). It is shown that the combination of GWO and GDA can improve the efficiency of GWO and avoid local optima. The results are compared with some metaheuristic algorithms. To test the model of 11 hospitals in Tehran, three different modes have been considered, and in each mode, 7... 

In recent years, the use of Deep Neural Networks in solving various problems has grown considerably. Possessing a large number of parameters, these networks have the ability to reconstruct complex functions and relations from large amounts of data and have been able to achieve the best results in a wide range of problems. But using these models comes with its own problems. These networks typically require considerable resources in order to run. This makes it inefficient or impossible to use them in systems with limited processing capabilities, e.g mobile phones. The existing approaches, e.g. the deployment of the model on a powerful server and network compression, have their own drawbacks... 

Characterizing cellular states in response to various disease conditions is an important issue which is addressed by different methods such as Large-scale gene expression profiling. One of the most important challenges in front of bioinformaticians is the loss of data because expression profiling is still very expensive. It is understood that profiling a group of selected genes could be enough for understanding all of the gene expression profile.In this research, we propose a fast method for estimation of the missing values inlow-rank matrices. We consider the highly correlated expression profiles as a low-rank matrix. Then, we used this new method in a proposed algorithm which will select... 

Ising model is a Markov Random Field (MRF) with binary random variables which has a vast literature in both theoretical and practical sides. In this thesis, we investigate two important statistical problems on this model. Learning the structure of MRFs has a long history and had a significant progress in the recent years. The goal of this problem is to find the independence graph of MRF using the samples generated from it. Specifically, we focus on the structure learning of ising models. Important algorithms for finding the structures had been reviewed. Additionally, we introduced information-theoretical and computational limitations of this problem. The second problem is community detection... 

Bulk tissue RNA-seq data has been widely used for investigating the transcriptome and analyzing it for different purposes. A single bulk sample of a heterogeneous population includes different cell-types each in different proportions. Bulk tissue RNA-seq measures the average expression level of genes across these cell types and does not account for cross-subject variation in cell-type compositions. Furthermore, biological signals might be masked by taking the average of gene expressions. Because of these reasons, bulk-RNA-seq is not suffcient for studying complex tissues. Knowing these cell-type compositions are important, because studying cell-specific changes in the transcriptome might be... 

The problem of separation of audio sources is one of the oldest issues raised in the field of audio processing, which has been studied for more than half a century. The main focus of recent research in this field has been on improving the sound quality resulting from the separation of sound sources with the help of deep neural networks. This is despite the fact that in most applications of audio source separation, such as the application of meeting transcription, we do not need the separated audio of people. Rather, we need a pipeline of converting overlapping speech to text, which, by receiving the audio in which several people have spoken, outputs the text spoken by the people present in... 

Histopathology images exhibit significant variations across different datasets, which can pose challenges for machine learning models. When a model is trained on a certain domain, its performance may severely deteriorate when applied to unseen domains, magnifying the necessity of out-of-distribution generalization. To address this problem, we present a novel framework that harnesses the strengths of self-supervised training to extract domain-invariant features, and combine it with domain transfer techniques. Our approach is built on the premise that pre-trained vision transformers, highly effective in natural image analysis, can be adapted for histopathology image analysis in a... 

Datasets extracted from gene expression microarrays contain information about the phenotypic behavior of organisms. Turning this information into knowledge, i.e. finding associative genes with a given phenotype, is a daunting task. This is due to the high dimensionality of the data as the number of features on a gene expression microarray is usually very large. Moreover, a phenotype may change the expression pattern of a set of genes rather than changing each gene’s expression independently. To tackle the second problem, integrating other sources of information such as Protein-Protein Interaction (PPI) networks is required. In this thesis, the PPI network extracted from the String database... 

Genome sequencing is one of the fundamental problems in today’s Biology. It has applications in significant problems such as finding association between an individual’s genome sequence and his phenotypes, discovering new genes, and finding evolutionary relations between organisms. Following the rapid advances in sequencing technologies and generation of a huge amount of short reads, efficient computational tools are needed for processing sequencing data. Shortness of reads is a factor that makes the task of reconstructing repetitive genomic regions complicated. In fact, the main challenge in both sequencing and resequencing problems is reconstruction of repeat regions. Common resequencing... 

Nowadays, many progresses in biology and medicine such as diagnosis of diseases and drug discoveries depend heavily on analyzing biological datasets collected from advanced machines. DNA Microarrays are amongst such machines applicable in measuring the expressions levels of thousand of genes and genotyping of a set of single nucleotide polymorphic sites to name a few. Compared to the more advanced Next Generation Sequencing (NGS) technology, the microarray platform produces lower quality of datasets. However, there has been tones of efforts to produce, process, and curate datasets from microarrays based on well designed protocols for sample preparation, hybridization, image processing, and... 

Isoforms are mRNAs that are produced from a same gene site in the phenomenon called Alternative Splicing. Studies have shown that more than 95% of multiexon genes in humans have undergone Alternative Splicing. Although there are few changes in mRNA sequence, They may have a systematic effect on cell function and regulation. It is widely reported that isoforms of a gene have distinct or even contrasting functions. Most studies have shown that alternative splicing plays a significant role in human health and disease. Despite the wide range of gene function studies, there is little information about isoforms’ functionalities. Recently, some computational methods based on Multiple Instance... 

Following dramatic changes after using deep learning method as a solution for Natural Language Processing tasks, Transformer architecture get popular. Based on that, then BERT Language model presented and get state-of-the-art as a solution for a lot of language processing tasks. It was a turning point in Natural Language Processing field. Also, in cross-lingual methods research line motivated by developing a common space for representation of language units, e.g. words, sentences, in more that one language, get some remarkable improvements. However, for languages distant from English such as Persian or Arabic the methods' performance was not clear. In this work, we performed some innovative... 

As latency reduction is a key objective of upcoming mobile standards such as 5G, control applications over such channels have become of high interest in recent years. In this thesis, we will consider the problem of control over fading channels modeled by finite-state Markov chains. We will combine communication and control theory aspects of the problem and study stability of linear discrete-time plants which communicate with the controller over such fading channels. Two different scenarios are assumed for channel state information availability at encoder side, and the sufficient condition of almost sure asymptotic stabilizability is derived for each scenario. We will then examine the... 

Development of DNA sequencing technologies in the recent years magnifies the need for computational tools in genomic data processing, and thus has attracted inten- sive research interest to this area. Among them, Genome-Wide Association Study (GWAS) refers to discovering of causal relationships among genetic sequences of living organisms and the macroscopic phenotypes present in their physiological structure. Chosen phenotypes for genomic association studies are mostly vulnerability or im- munity to common genetic diseases. Conventional methods in GWAS consists of statistical hypothesis testing algorithms in case/control approaches; Most of which are based upon single-locus analysis and...