This dissertation is concerned with the application of mathematical modeling and statistical signal processing into the rapidly expanding fields of proteomics and genomics. The research is guided by a translational goal which drives the problem formalization and experimental design, and leads to optimization, prediction and control of the underlying system. The dissertation is comprised of three interconnected subjects. In the first part of the dissertation, two Bayesian peptide detection algorithms are proposed to optimize the feature extraction step, which is the most fundamental step in mass spectrometry-based proteomics. The algorithms are designed to tackle data processing challenges that are not satisfactorily addressed by existing methods. In contrast to most existing methods, the proposed algorithms perform deisotoping and deconvolution of mass spectra simultaneously, which enables better identification of weak peptide signals. Unlike greedy template-matching algorithms, the proposed methods have the capability to handle complex spectra where features overlap. The proposed methods achieve better sensitivity and accuracy compared to many popular software packages such as msInspect. In the second part of the dissertation, we consider modeling and assessing the entire mass spectrometry-based proteomic data analysis pipeline. Different modules are identified and analyzed, resulting in a framework that captures key factors in system performance. The effects of various model parameters on protein identification rates and quantification errors, differential expression results, and classification performance are examined. The proposed pipeline model can be used to aid experimental design, pinpoint critical bottlenecks, optimize the work flow, and predict biomarker discovery results. Finally, the same system methodology is extended to analyze the work flow in DNA microarray experiments. A model-based approach is developed to explore the relationship among microarray data properties, missing value imputation, and sample classification in a complicated data analysis pipeline. The situations when it is suitable to apply missing value imputation are identified and recommendations regarding imputation are provided. In addition, a missing value rate-related peaking phenomenon is uncovered.

Biomarkers are biological characteristics that can be observed or measured during disease conditions, and compared to the healthy state. Biomarkers have been used in medical history to study disease progression, to develop drugs, or to predict drug efficacy. However, in complex diseases such as in cancer, biomarkers vary tremendously among patients and disease stages. Cell surface receptors, proteins that are located at the cell surface and deliver external signals into the cell, are a significant group of easily-detectable biomarkers. Along with the detection of particular biomarkers related to a disease, extensive characterization of expression patterns is necessary to optimize their application. Therefore, we designed a technique to imprint or capture the expression pattern of these receptors on silver nanoparticles. We incorporated branched molecules that can simultaneously bind to the target receptors and the nanoparticle surface. To develop the technique, we used melanocortin receptor 1 (MC1R), a receptor present at high levels on the surface of melanoma cells, as a test system. We determined optimum binding of this molecule in an established melanoma cell line, WM-266-4. We also synthesized a labeled molecule that was used to estimate the number of MC1R proteins on these cells. These studies indicate that this might be a promising approach for developing sensitive and cost-effective tools to characterize cell surface receptors in studying complex diseases and cell mechanisms. / MS / Biomarkers are biological characteristics that can be observed or measured during disease conditions, and compared to the healthy state (e.g. grades of fever during infection). Biomarkers have been used in medical history to study disease progression, to develop drugs, or to predict drug efficacy. However, in complex diseases such as in cancer, biomarkers vary tremendously among patients and disease stages. Cell surface receptors, proteins that are located at the cell surface and deliver external signals into the cell, are a significant group of easily-detectable biomarkers. Along with the detection of particular biomarkers related to a disease, extensive characterization of expression patterns is necessary to optimize their application. Therefore, we designed a technique to imprint or capture the expression pattern of these receptors on silver nanoparticles. We incorporated branched molecules that can simultaneously bind to the target receptors and the nanoparticle surface. To develop the technique, we used melanocortin receptor 1 (MC1R), a receptor present at high levels on the surface of melanoma cells, as a test system. We determined optimum binding of this molecule in an established melanoma cell line, WM-266-4. We also synthesized a labeled molecule that was used to estimate the number of MC1R proteins on these cells. These studies indicate that this might be a promising approach for developing sensitive and cost-effective tools to characterize cell surface receptors in studying complex diseases and cell mechanisms.

Luminescent semiconductor quantum dots (QDs) are extensively researched for use in biological applications. They have unique optical and physical properties that make them excellent candidates to replace conventional organic dyes for cellular labeling, multiplexing, nucleic acid detection, and as generalized probes. The primary focus of this dissertation was to utilize quantum dots for improvement in immunoassays. Specifically, atherosclerosis biomarkers were detected simultaneously in an effort to demonstrate advances in early detection diagnostics. Quantum dot-antibody bioconjugates were prepared by encapsulation into mesoporous silica and functionalized with thiol and amine groups to enable bioconjugation. Functionalization of the mesoporous silica quantum dot composites facilitated biocompatibility for use with biological buffers in immunoassays. These bioconjugates were used in a sandwich immunoassay to detect atherosclerosis biomarkers IL-15 and MCP-1. Sandwich assays employ capture antibodies immobilized onto a well plate to bind as much of the antigen as possible. The capture antibodies increased binding by at least 4 times the amount of antigen bound to the surface of a direct detection assay. The sandwich immunoassay was able to detect 1 pg/mL of IL-15 and 50 pg/mL of MCP-1 biomarkers. Human serum albumin nanoparticles (HSAPs) were synthesized via a desolvation and crosslinking method. Human serum albumin is a versatile protein being used in a variety of applications. Quantum dots were loaded into HSAPs as potential detection probes for immunoassays. Efficient loading was not achieved, and the assay was unable to improve current detection limits. Controlled release studies were explored using HSAPs loaded with superparamagnetic iron oxide nanoparticles and a fluorescent drug analog. Exposure to a magnetic field resulted in degradation of the HSAPs. The fluorophore was released and measured to examine how cancer drugs might be controlled through a magnetic field. Gold nanorods and an anticancer drug, Sorafenib, were also encapsulated into HSAPs for treatment of renal cell carcinoma in vivo. Laser irradiation treatment combined with Sorafenib resulted in 100% tumor necrosis and total elimination of any viable tumor present. HSAPs have demonstrated remarkable potential as drug delivery nanocarriers.

Over the past decade, solid-state nanopores have emerged as a versatile tool for the detection and characterization of single molecules, showing great promise in the field of personalized medicine as diagnostic and genotyping platforms. While solid-state nanopores offer increased durability and functionality over a wider range of experimental conditions compared to their biological counterparts, reliable fabrication of low-noise solid-state nanopores remains a challenge. In this thesis, a methodology for treating nanopores using high electric fields in an automated fashion by applying short (0.1-2 s) pulses of 6-10 V is presented which drastically improves the yield of nanopores that can be used for molecular recognition studies. In particular, this technique allows for sub-nanometer control over nanopore size under experimental conditions, facilitates complete wetting of nanopores, reduces noise by up to three orders of magnitude and rejuvenates used pores for further experimentation. This improvement in fabrication yield (over 90%) ultimately makes nanopore-based sensing more efficient, cost-effective and accessible. Tuning size using high electric fields facilitates nanopore fabrication and improves functionality for single-molecule experiments. Here, the use of nanopores for the detection of DNA-protein complexes is examined. As proof-of-concept, neutravidin bound to double-stranded DNA is used as a model complex. The creation of the DNA-neutravidin complex using polymerase chain reaction with biotinylated primers and subsequent purification and multiplex creation is discussed. Finally, an outlook for extending this scheme for the identification of proteins in a sample based on translocation signatures is presented which could be implemented in a portable lab-on-a-chip device for the rapid detection of disease biomarkers.

Over the past decade, solid-state nanopores have emerged as a versatile tool for the detection and characterization of single molecules, showing great promise in the field of personalized medicine as diagnostic and genotyping platforms. While solid-state nanopores offer increased durability and functionality over a wider range of experimental conditions compared to their biological counterparts, reliable fabrication of low-noise solid-state nanopores remains a challenge. In this thesis, a methodology for treating nanopores using high electric fields in an automated fashion by applying short (0.1-2 s) pulses of 6-10 V is presented which drastically improves the yield of nanopores that can be used for molecular recognition studies. In particular, this technique allows for sub-nanometer control over nanopore size under experimental conditions, facilitates complete wetting of nanopores, reduces noise by up to three orders of magnitude and rejuvenates used pores for further experimentation. This improvement in fabrication yield (over 90%) ultimately makes nanopore-based sensing more efficient, cost-effective and accessible. Tuning size using high electric fields facilitates nanopore fabrication and improves functionality for single-molecule experiments. Here, the use of nanopores for the detection of DNA-protein complexes is examined. As proof-of-concept, neutravidin bound to double-stranded DNA is used as a model complex. The creation of the DNA-neutravidin complex using polymerase chain reaction with biotinylated primers and subsequent purification and multiplex creation is discussed. Finally, an outlook for extending this scheme for the identification of proteins in a sample based on translocation signatures is presented which could be implemented in a portable lab-on-a-chip device for the rapid detection of disease biomarkers.

Over the past decade, solid-state nanopores have emerged as a versatile tool for the detection and characterization of single molecules, showing great promise in the field of personalized medicine as diagnostic and genotyping platforms. While solid-state nanopores offer increased durability and functionality over a wider range of experimental conditions compared to their biological counterparts, reliable fabrication of low-noise solid-state nanopores remains a challenge. In this thesis, a methodology for treating nanopores using high electric fields in an automated fashion by applying short (0.1-2 s) pulses of 6-10 V is presented which drastically improves the yield of nanopores that can be used for molecular recognition studies. In particular, this technique allows for sub-nanometer control over nanopore size under experimental conditions, facilitates complete wetting of nanopores, reduces noise by up to three orders of magnitude and rejuvenates used pores for further experimentation. This improvement in fabrication yield (over 90%) ultimately makes nanopore-based sensing more efficient, cost-effective and accessible. Tuning size using high electric fields facilitates nanopore fabrication and improves functionality for single-molecule experiments. Here, the use of nanopores for the detection of DNA-protein complexes is examined. As proof-of-concept, neutravidin bound to double-stranded DNA is used as a model complex. The creation of the DNA-neutravidin complex using polymerase chain reaction with biotinylated primers and subsequent purification and multiplex creation is discussed. Finally, an outlook for extending this scheme for the identification of proteins in a sample based on translocation signatures is presented which could be implemented in a portable lab-on-a-chip device for the rapid detection of disease biomarkers.

Diseases are often diagnosed by detection of disease-specific biomarkers in fluid samples. However, many state-of-the-art detection methods require a lab with complex machinery, trained operators, and/or lengthy analysis time. In contrast, point-of-care (POC) devices are brought to the patient's location, they are easy to use, and results are obtained almost immediately. Many current POC devices are too difficult to be used without a skilled assistant, and although many are able to detect analytes above a threshold value, they give little or no quantitative information. This work presents the development of polymer-based microfluidic devices capable of sensing and quantifying biomarkers in fluid samples in a straightforward manner using a novel biomarker assay termed "flow valve diagnostics". In this assay, an antibody-modified polydimethylsiloxane (PDMS) microchannel constricts due to the binding force between antibodies and antigens, stopping fluid flow. The flow distance is measured and correlated to antigen concentration. This detection method is an improvement over other methods because it is an innovative, non-instrumented, label-free, easy-to-use approach. These devices are small, portable, disposable, inexpensive, and thus ideal for use in POC testing. I have successfully fabricated flow valve devices with standard micromachining techniques, including photolithography, replica molding with PDMS, and plasma oxidation. Following fabrication, I compared two methods for attaching receptor biomolecules (e.g., antibodies) to the microchannel surfaces: non-specific adsorption and silanization with 3-glycidoxytrimethoxypropylsilane (GOPS). I used laser-induced fluorescence to determine that silanization with GOPS was the better method for biomolecule attachment. Finally, I tested antibody-modified flow valve devices with target antigens to determine if the antibody/antigen binding force was strong enough to cause channel pinching and flow stoppage. By modifying the device design and using higher antigen concentrations, I was able to show that flow valve devices can detect antigens in a concentration-dependent manner. Future work to improve the device design and to modify and test these devices with different receptor/target pairs will bring flow valve diagnostics closer to becoming a valuable asset in biomarker detection and POC testing.

Droplet microfluidics is an emerging technology for analysis of single cells and biomolecules at high throughput. The controlled encapsulation of particles along with the surrounding microenvironment in discrete droplets, which acts as miniaturized reaction vessels, allows millions of particles to be screened in parallel. By utilizing the unit operations developed to generate, manipulate and analyze droplets, this technology platform has been used to miniaturize a wide range of complex biological assays including, but not limited to, directed evolution, rare cell detection, single cell transcriptomics, rare mutation detection and drug screening. The aim of this thesis is to develop droplet microfluidics based methods for analysis of single cells and nucleic acids. In Paper I, a method for time-series analysis of mammalian cells, using automated fluorescence microscopy and image analysis technique is presented. The cell-containing droplets were trapped on-chip and imaged continuously to assess the viability of hundreds of isolated individual cells over time. This method can be used for studying the dynamic behavior of cells. In Paper II, the influence of droplet size on cell division and viability of mammalian cell factories during cultivation in droplets is presented. The ability to achieve continuous cell division in droplets will enable development of mammalian cell factory screening assays in droplets. In Paper III, a workflow for detecting the outcome of droplet PCR assay using fluorescently color-coded beads is presented. This workflow was used to detect the presence of DNA biomarkers associated with poultry pathogens in a sample. The use of color-coded detection beads will help to improve the scalability of the detection panel, to detect multiple targets in a sample. In Paper IV, a novel unit operation for label-free enrichment of particles in droplets using acoustophoresis is presented. This technique will be useful for developing droplet-based assays that require label-free enrichment of cells/particles and removal of droplet content. In general, droplet microfluidics has proven to be a versatile tool for biological analysis. In the years to come, droplet microfluidics could potentially be used to improve clinical diagnostics and bio-based production processes. / <p>QC 20160926</p>

Medical imagery is increasingly evolving towards higher resolution and throughput. The increasing volume of data and the usage of multiple and often novel imaging modalities necessitates the use of mathematical and computational techniques for quicker, more accurate and more robust analysis of medical imagery. The fields of computer vision and machine learning provide a rich set of techniques that are useful in medical image analysis, in tasks ranging from segmentation to classification and population analysis, notably by integrating the qualitative knowledge of experts in anatomy and the pathologies of various disorders and making it applicable to the analysis of medical imagery going forward. The object of the proposed research is exactly to explore various computer vision and machine learning methods with a view to the improved analysis of multiple modalities of brain and cardiac imagery, towards enabling the clinical goals of studying schizophrenia, brain tumors (meningiomas and gliomas in particular) and cardiovascular disorders. In the first project, a framework is proposed for the segmentation of tubular, branched anatomical structures. The framework uses the tubular surface model which yields computational advantages and further incorporates a novel automatic branch detection algorithm. It is successfully applied to the segmentation of neural fiber bundles and blood vessels. In the second project, a novel population analysis framework is built using the shape model proposed as part of the first project. This framework is applied to the analysis of neural fiber bundles towards the detection and understanding of schizophrenia. In the third and final project, the use of mass spectrometry imaging for the analysis of brain tumors is motivated on two fronts, towards the offline classification analysis of the data, as well as the end application of intraoperative detection of tumor boundaries. SVMs are applied for the classification of gliomas into one of four subtypes towards application in building appropriate treatment plans, and multiple statistical measures are studied with a view to feature extraction (or biomarker detection). The problem of intraoperative tumor boundary detection is formulated as a detection of local minima of the spatial map of tumor cell concentration which in turn is modeled as a function of the mass spectra, via regression techniques.

Point-of-care testing (POCT) is defined as medical testing at or near the site of patient care and has become a critical component of the diagnostic industry. POCT has many advantages over tests in centralized laboratories including small reagent volumes, small size, rapid turnaround time, cost-effectiveness, low power consumption and functional integration of multiple devices. Paper-based POCT sensors are a new alternative technology for fabricating simple, low-cost, portable and disposable analytical devices for clinical diagnosis. The focus of this dissertation was to develop simple, rapid and low cost paper-based POCT sensors with high sensitivity and portability for disease biomarker detection. Lateral flow strips (LFS) were used as the basic platform as it provides several key advantages such as simplicity, fast response time, on site and cost-effectiveness, and it can be used to detect specific substances including small molecules, large proteins and even whole pathogens, in a sample by immunological reactions. Earlier designs of paper strips lacked the quantitative information of the analyte concentration and could only provide single analyte detection at a time. In this study, a series of modifications were made to upgrade the platform to compensate for these limitations. First, we developed a gold nanoparticle based LFS for qualitative colorimetrical detection of bladder cancer related biomarkers in standard solutions and in urine samples. Second, by incorporating an image processing program “ImageJ”, a semi-quantitative LFS platform was established. The capability of the strip was evaluated by testing a small DNA oxidative damage biomarker in urine and cell culture models. Third, we combined the electrochemical method and colorimetrical method for quantitative biomarker detection. Finally, we integrated a commercialized blood glucose meter to quantitatively detection of two non-glucose biomarkers by converting their signals to that of glucose. The upgraded sensor could provide a noninvasive, rapid, visual, quantitative and convenient detection platform for various disease biomarkers. In addition, this platform does not require expensive equipments or trained personnel, deeming it suitable for use as a simple, economical and portable field kit for on-site biomarker monitoring in a variety of clinical settings.