Formal Concept Analysis for Image Classification and Machine Learning Models for Anti-CRISPR Protein Discovery in Bioinformatics

Aug 5, 2026

Upload by Ivy

This dissertation advances two bioinformatics needs: improving transparency in medical image analysis and accelerating discovery of Anti-CRISPR (Acr) proteins for more precise CRISPR-Cas gene editing. It addresses limitations of CNN interpretability methods by using Formal Concept Analysis (FCA) to model the relationship between abstract features and class labels across datasets. FCA-based performance is validated on MNIST and histopathological datasets. The study also leverages genomic context to mitigate Acr data scarcity and build machine learning models for new Anti-CRISPR discovery, while 3D structure analysis supports protein classification by Acr type and CRISPR-Cas system.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages59

Reading Time2 h 29 min

AccessFree to download

Contents

Introduction

Image Analysis

Discovery of Anti-CRISPR proteins

Literature Review

Image Classification Models

Formal Concept Analysis (FCA)

Bioinformatics Tools for Discovery of Anti-CRISPR Proteins

Formal Concept Analysis (FCA) Classifier

Application of FCA classifier on MNIST Dataset

University of Nebraska-Lincoln DigitalCommons@University of Nebraska-Lincoln | Dissertations and Doctoral Documents from University of Nebraska-Lincoln, 2024– | Graduate Studies | | --- | --- | | 12-5-2023
Formal Concept Analysis for Image Classification and Machine Learning Models for Anti-CRISPR Protein Discovery in Bioinformatics
Minal Khatri
University of Nebraska-Lincoln, [khatri.cs16@gmail.com](khatri.cs16@gmail.com)
Follow this and additional works at: [https://digitalcommons.unl.edu/dissunl](https://digitalcommons.unl.edu/dissunl)
Part of the Computer Sciences Commons | | Recommended Citation Khatri, Minal, "Formal Concept Analysis for Image Classification and Machine Learning Models for AntiCRISPR Protein Discovery in Bioinformatics" (2023) . Dissertations and Doctoral Documents from University of Nebraska-Lincoln, 2024–. 47. [https://digitalcommons.unl.edu/dissunl/47](https://digitalcommons.unl.edu/dissunl/47) This Dissertation is brought to you for free and open access by the Graduate Studies at DigitalCommons@University of Nebraska-Lincoln. It has been accepted for inclusion in Dissertations and Doctoral Documents from University of Nebraska-Lincoln, 2024– by an authorized administrator of DigitalCommons@University of Nebraska-Lincoln. FORMAL CONCEPT ANALYSIS FOR IMAGE CLASSIFICATION AND MACHINE LEARNING MODELS FOR ANTI-CRISPR PROTEIN DISCOVERY IN BIOINFORMATICS by Minal Khatri A DISSERTATION Presented to the Faculty of The Graduate College at the University of Nebraska In Partial Fulfillment of Requirements For the Degree of Doctor of Philosophy Major: Computer Science Under the Supervision of Professors Jitender S. Deogun & Yanbin Yin Lincoln, Nebraska December, 2023 FORMAL CONCEPT ANALYSIS FOR IMAGE CLASSIFICATION AND MACHINE LEARNING MODELS FOR ANTI-CRISPR PROTEIN DISCOVERY IN BIOINFORMATICS Minal Khatri, Ph.D. University of Nebraska, 2023 Adviser: Professors Jitender S. Deogun & Yanbin Yin This study investigates two critical areas in bioinformatics: enhancing transparency in medical image analysis and advancing the discovery of Anti-CRISPR (Acr) proteins, which have potential in developing more precise and controlled CRISPR-Cas gene editing tools. While CNN’s are increasingly applied in critical fields like medical diagnosis, understanding their decision-making process remains a challenge. Although visualization techniques like Saliency maps offer insights into CNN’s decision-making for individual images, they do not explicitly establish a relationship between the high-level features learned by CNN’sand the class labels across dataset. To bridge this gap, Formal Concept Analysis (FCA) framework is leveraged as a image classification model, establishing a novel method for understanding the relationship between abstract features and class labels in medical imaging. The model’s performance is validated across a range from the simpler MNIST dataset to more complex histopathological image datasets like Warwick-QUand BreakHIS. Simultaneously, the study explores the genomic context of Acr genes and the 3D structure of Acr proteins for Acr discovery, which are not extensively explored in current bioinformatics tools. By leveraging genomic context, we overcome data scarcity and develop a machine learning model capable of discovering new Anti-CRISPRs. Additionally, the 3D structure analysis aids in developing machine learning classifiersto classify proteins by Acr type and CRISPR-Cas systems. Overall, this research makes significant contributions to the field of bioinformatics, by developing robust methodologies, enhancing our understanding of medical image analysis and advances Acr protein discovery. iii ACKNOWLEDGMENTS I extend my deepest gratitude to my advisors, Dr. Jitender Deogun and Dr. Yanbin Yin. Their exceptional expertise, understanding, and patience have profoundly enriched my graduate experience. I am immensely thankful for their genero

FAQs

How does the dissertation improve transparency in medical image classification beyond saliency maps?

It uses a Formal Concept Analysis (FCA) framework to explicitly connect abstract features learned from images to class labels across datasets, rather than focusing on per-image visual explanations.

Which datasets are used to validate the FCA image classification approach?

The performance is validated on MNIST and on more complex histopathological image datasets such as Warwick-QUand BreakHIS.

How does the research address data scarcity in Anti-CRISPR (Acr) protein discovery?

It leverages the genomic context of Acr genes to enable machine learning models that can discover new Anti-CRISPR proteins despite limited labeled data.

Aug 5, 2026

Upload by Ivy