Publications with Fulltext

Permanent URI for this collectionhttps://hdl.handle.net/20.500.14288/6

Browse

Filters

Advanced Search

Filter by

Settings

Department: search.filters.department.Department of Computer EngineeringQuartile: search.filters.quartile.Q1

Search Results

Now showing 1 - 10 of 24
  • Thumbnail Image
    PublicationOpen Access
    The noisy channel mode for unsupervised word sense disambiguation
    (Massachusetts Institute of Technology (MIT) Press, 2010) Department of Computer Engineering; Yüret, Deniz; Yatbaz, Mehmet Ali; Faculty Member; PhD Student; Department of Computer Engineering; College of Engineering; 179996; 192506
    We introduce a generative probabilistic model, the noisy channel model, for unsupervised word sense disambiguation. In our model, each context C is modeled as a distinct channel through which the speaker intends to transmit a particular meaning S using a possibly ambiguous word W. To reconstruct the intended meaning the hearer uses the distribution of possible meanings in the given context P(S|C) and possible words that can express each meaning P(W|S). We assume P(W|S) is independent of the context and estimate it using WordNet sense frequencies. The main problem of unsupervised WSD is estimating context-dependent P(S|C) without access to any sense-tagged text. We show one way to solve this problem using a statistical language model based on large amounts of untagged text. Our model uses coarse-grained semantic classes for S internally and we explore the effect of using different levels of granularity on WSD performance. The system outputs fine-grained senses for evaluation, and its performance on noun disambiguation is better than most previously reported unsupervised systems and close to the best supervised systems.
  • Thumbnail Image
    PublicationOpen Access
    HotRegion: a database of predicted hot spot clusters
    (Oxford University Press (OUP), 2012) N/A; Department of Computer Engineering; Department of Chemical and Biological Engineering; Çukuroğlu, Engin; Gürsoy, Attila; Keskin, Özlem; PhD Student; Faculty Member; Department of Computer Engineering; Department of Chemical and Biological Engineering; Graduate School of Sciences and Engineering; College of Engineering; N/A; 8745; 26605
    Hot spots are energetically important residues at protein interfaces and they are not randomly distributed across the interface but rather clustered. These clustered hot spots form hot regions. Hot regions are important for the stability of protein complexes, as well as providing specificity to binding sites. We propose a database called HotRegion, which provides the hot region information of the interfaces by using predicted hot spot residues, and structural properties of these interface residues such as pair potentials of interface residues, accessible surface area (ASA) and relative ASA values of interface residues of both monomer and complex forms of proteins. Also, the 3D visualization of the interface and interactions among hot spot residues are provided.
  • Thumbnail Image
    PublicationOpen Access
    PRISM: a web server and repository for prediction of protein-protein interactions and modeling their 3D complexes
    (Oxford University Press (OUP), 2014) Nussinov, Ruth; Department of Computer Engineering; Department of Chemical and Biological Engineering; Department of Computer Engineering; Başpınar, Alper; Çukuroğlu, Engin; Keskin, Özlem; Gürsoy, Attila; PhD Student; Faculty Member; Department of Chemical and Biological Engineering; Department of Computer Engineering; Graduate School of Sciences and Engineering; College of Engineering; N/A; N/A; 26605; 8745
    The PRISM web server enables fast and accurate prediction of protein-protein interactions (PPIs). The prediction algorithm is knowledge-based. It combines structural similarity and accounts for evolutionary conservation in the template interfaces. The predicted models are stored in its repository. Given two protein structures, PRISM will provide a structural model of their complex if a matching template interface is available. Users can download the complex structure, retrieve the interface residues and visualize the complex model.
  • Thumbnail Image
    PublicationOpen Access
    Usable optimistic fair exchange
    (Elsevier, 2012) Lysyanskaya, A.; Department of Computer Engineering; Küpçü, Alptekin; Faculty Member; Department of Computer Engineering; College of Engineering; 168060
    Fairly exchanging digital content is an everyday problem. It has been shown that fair exchange cannot be achieved without a trusted third party (called the Arbiter). Yet, even with a trusted party, it is still non-trivial to come up with an efficient solution, especially one that can be used in a p2p file sharing system with a high volume of data exchanged. We provide an efficient optimistic fair exchange mechanism for bartering digital files, where receiving a payment in return for a file (buying) is also considered fair. The exchange is optimistic, removing the need for the Arbiter's involvement unless a dispute occurs. While the previous solutions employ costly cryptographic primitives for every file or block exchanged, our protocol employs them only once per peer, therefore achieving an O(n) efficiency improvement when n blocks are exchanged between two peers. Our protocol uses very efficient cryptography, making it perfectly suitable for a p-2-p file sharing system where tens of peers exchange thousands of blocks and they do not know beforehand which ones they will end up exchanging. Therefore, our system yields up to one-to-two orders of magnitude improvement in terms of both computation and communication (40s vs. 42 min, 1.6 MB vs. 200 MB). Thus, for the first time, a provably secure (and privacy-respecting when payments are made using e-cash) fair exchange protocol can be used in real bartering applications (e.g., BitTorrent) [14] without sacrificing performance. (C) 2011 Elsevier B.V. All rights reserved.
  • Thumbnail Image
    PublicationOpen Access
    PRISM: protein interactions by structural matching
    (Oxford University Press (OUP), 2005) Nussinov, Ruth; Department of Computer Engineering; Department of Chemical and Biological Engineering; Department of Computer Engineering; Öğmen, Utkan; Keskin, Özlem; Aytuna, Ali Selim; Gürsoy, Attila; Faculty Member; Department of Chemical and Biological Engineering; Department of Computer Engineering; Graduate School of Sciences and Engineering; College of Engineering; N/A; 26605; N/A; 8745
    Prism (https://gordion.hpc.eng.ku.edu.tr/prism) is a website for protein interface analysis and prediction of putative protein-protein interactions. It is composed of a database holding protein interface structures derived from the Protein Data Bank (PDB). The server also includes summary information about related proteins and an interactive protein interface viewer. A list of putative protein-protein interactions obtained by running our prediction algorithm can also be accessed. These results are applied to a set of protein structures obtained from the PDB at the time of algorithm execution (January 2004). Users can browse through the non-redundant dataset of representative interfaces on which the prediction algorithm depends, retrieve the list of similar structures to these interfaces or see the results of interaction predictions for a particular protein. Another service provided is interactive prediction. This is done by running the algorithm for user input structures.
  • Thumbnail Image
    PublicationOpen Access
    AffectON: incorporating affect into dialog generation
    (Institute of Electrical and Electronics Engineers (IEEE), 2020) Bucinca, Zana; Department of Computer Engineering; Yemez, Yücel; Erzin, Engin; Sezgin, Tevfik Metin; Faculty Member; Faculty Member; Faculty Member; Department of Computer Engineering; College of Engineering; 107907; 34503; 18632
    Due to its expressivity, natural language is paramount for explicit and implicit affective state communication among humans. The same linguistic inquiry (e.g. How are you ?) might induce responses with different affects depending on the affective state of the conversational partner(s) and the context of the conversation. Yet, most dialog systems do not consider affect as constitutive aspect of response generation. In this paper, we introduce AffectON, an approach for generating affective responses during inference. For generating language in a targeted affect, our approach leverages a probabilistic language model and an affective space. AffectON is language model agnostic, since it can work with probabilities generated by any language model (e.g., sequence-to-sequence models, neural language models, n-grams). Hence, it can be employed for both affective dialog and affective language generation. We experimented with affective dialog generation and evaluated the generated text objectively and subjectively. For the subjective part of the evaluation, we designed a custom user interface for rating and provided recommendations for the design of such interfaces. The results, both subjective and objective demonstrate that our approach is successful in pulling the generated language toward the targeted affect, with little sacrifice in syntactic coherence.
  • Thumbnail Image
    PublicationOpen Access
    Modeling protein assemblies in the proteome
    (American Society for Biochemistry and Molecular Biology (ASBMB), 2014) Nussinov, Ruth; Department of Chemical and Biological Engineering; Department of Computer Engineering; Kuzu, Güray; Keskin, Özlem; Gürsoy, Attila; Faculty Member; Department of Chemical and Biological Engineering; Department of Computer Engineering; College of Engineering; N/A; 26605; 8745
    Most (if not all) proteins function when associated in multimolecular assemblies. Attaining the structures of protein assemblies at the atomic scale is an important aim of structural biology. Experimentally, structures are increasingly available, and computations can help bridge the resolution gap between high- and low-resolution scales. Existing computational methods have made substantial progress toward this aim; however, current approaches are still limited. Some involve manual adjustment of experimental data; some are automated docking methods, which are computationally expensive and not applicable to large-scale proteome studies; and still others exploit the symmetry of the complexes and thus are not applicable to nonsymmetrical complexes. Our study aims to take steps toward overcoming these limitations. We have developed a strategy for the construction of protein assemblies computationally based on binary interactions predicted by a motif-based protein interaction prediction tool, PRISM (Protein Interactions by Structural Matching). Previously, we have shown its power in predicting pairwise interactions. Here we take a step toward multimolecular assemblies, reflecting the more prevalent cellular scenarios. With this method we are able to construct homo-/hetero-complexes and symmetric/asymmetric complexes without a limitation on the number of components. The method considers conformational changes and is applicable to large-scale studies. We also exploit electron microscopy density maps to select a solution from among the predictions. Here we present the method, illustrate its results, and highlight its current limitations.
  • Thumbnail Image
    PublicationOpen Access
    Artificial intelligence approaches to human-microbiome protein-protein interactions
    (Elsevier, 2022) Lim, Hansaim; Tsai, Chung-Jung; Nussinov, Ruth; Department of Computer Engineering; Department of Chemical and Biological Engineering; Gürsoy, Attila; Keskin, Özlem; Faculty Member; Department of Computer Engineering; Department of Chemical and Biological Engineering; Koç University Research Center for Translational Medicine (KUTTAM) / Koç Üniversitesi Translasyonel Tıp Araştırma Merkezi (KUTTAM); College of Engineering; Graduate School of Sciences and Engineering; 8745; 26605; N/A
    Host-microbiome interactions play significant roles in human health and disease. Artificial intelligence approaches have been developed to better understand and predict the molecular interplay between the host and its microbiome. Here, we review recent advancements in computational methods to predict microbial effects on human cells with a special focus on protein–protein interactions. We categorize recent methods from traditional ones to more recent deep learning methods, followed by several challenges and potential solutions in structure-based approaches. This review serves as a brief guide to the current status and future directions in the field.
  • Thumbnail Image
    PublicationOpen Access
    Edge intelligence for empowering IoT-based healthcare systems
    (Institute of Electrical and Electronics Engineers (IEEE), 2021) Aloqaily, Moayad; Guizani, Mohsen; Department of Computer Engineering; Özkasap, Öznur; Hayyolalam, Vahideh; Faculty Member; Department of Computer Engineering; College of Engineering; Graduate School of Sciences and Engineering; 113507; N/A
    The demand for real-time, affordable, and efficient smart healthcare services is increasing exponentially due to the technological revolution and burst of population. To meet the increasing demands on this critical infrastructure, there is a need for intelligent methods to cope with the existing related challenges. In this regard, edge computing technology can reduce latency and energy consumption by moving processes closer to the data sources in comparison to the traditional centralized cloud and IoT-based healthcare systems. In addition, by bringing automated insights into the smart healthcare systems, artificial intelligence (AI) provides the possibility of detecting and predicting high-risk diseases in advance, decreasing medical costs for patients, and offering efficient treatments. The objective of this article is to highlight the benefits of the adoption of the edge intelligent technology, along with the use of AI in smart healthcare systems. Moreover, a novel smart healthcare model is proposed to boost the utilization of AI and edge technology in smart healthcare systems. Additionally, we discuss potential challenges and future research directions arising when integrating these different technologies.
  • Thumbnail Image
    PublicationOpen Access
    3D spatial organization and network-guided comparison of mutation profiles in Glioblastoma reveals similarities across patients
    (Public Library of Science, 2019) Dinçer, Cansu; Kaya, Tuğba; Tunçbağ, Nurcan; Department of Chemical and Biological Engineering; Department of Computer Engineering; Keskin, Özlem; Gürsoy, Attila; Faculty Member; Department of Chemical and Biological Engineering; Department of Computer Engineering; Koç University Research Center for Translational Medicine (KUTTAM) / Koç Üniversitesi Translasyonel Tıp Araştırma Merkezi (KUTTAM); College of Engineering; 26605; 8745
    Glioblastoma multiforme (GBM) is the most aggressive type of brain tumor. Molecular heterogeneity is a hallmark of GBM tumors that is a barrier in developing treatment strategies. In this study, we used the nonsynonymous mutations of GBM tumors deposited in The Cancer Genome Atlas (TCGA) and applied a systems level approach based on biophysical characteristics of mutations and their organization in patient-specific subnetworks to reduce inter-patient heterogeneity and to gain potential clinically relevant insights. Approximately 10% of the mutations are located in "patches" which are defined as the set of residues spatially in close proximity that are mutated across multiple patients. Grouping mutations as 3D patches reduces the heterogeneity across patients. There are multiple patches that are relatively small in oncogenes, whereas there are a small number of very large patches in tumor suppressors. Additionally, different patches in the same protein are often located at different domains that can mediate different functions. We stratified the patients into five groups based on their potentially affected pathways, revealed from the patient-specific subnetworks. These subnetworks were constructed by integrating mutation profiles of the patients with the interactome data. Network-guided clustering showed significant association between each group and patient survival (P-value = 0.0408). Also, each group carries a set of signature 3D mutation patches that affect predominant pathways. We integrated drug sensitivity data of GBM cell lines with the mutation patches and the patient groups to analyze the therapeutic outcome of these patches. We found that Pazopanib might be effective in Group 3 by targeting CSF1R. Additionally, inhibiting ATM that is a mediator of PTEN phosphorylation may be ineffective in Group 2. We believe that from mutations to networks and eventually to clinical and therapeutic data, this study provides a novel perspective in the network-guided precision medicine.