Innovative Projects Realized

Computational design for improving the affinity of peptide ligands for carbohydrate-binding

Diarrhoeal diseases pose a serious global problem, especially in the developing world. Shigella flexneri Y, the most deadly species of Shigella, is endemic in most developing countries. An antibody, SYA/J6, against Shigella flexneri Y was developed. Studies have shown that MDWNMHAA is a weakly immunogenic peptide against the antibody SYA/J6. The research team at Zymeoworks Inc., a computational biotechnology company, is hoping to improve the binding of MDWNMHAA to SYA/J6 in order to generate a more effective immune response. This could lead to a vaccine for Shigella flexneri Y. The MITACS Accelerate intern, a Ph.D. student, will modify the structure of MDWNMHAA via mutagenesis, which involves the replacement of one amino acid for another. She will then evaluate and optimize the binding properties of the newly designed molecules using molecular dynamics simulations and other computational tools. Guidance and resources for these studies will be provided by Zymeworks, Inc.

Approximate Algorithms for Resequencing of Human Genome using High Throughput Short

Human genetic variation has been traditionally studied at the single nucleotide polymorphism (SNP) level. It is clear that, human genetic variation extends beyond single nucleotide polymorphism. New projects aimed at identifying structural variation have been initiated through the use of high throughput sequencing technologies. Although the new sequencing technologies produce short reads with respect to Sanger sequencing experiments, the focus of search for differences between a sample and its reference is moving from single based differences to structural variations. The existing analysis methods concentrate on the mapping of individual reads and do not seek confirmation from other reads in the experimental data. As a consequence, they miss certain kinds of structural variations even if they are well‐represented in the data set. The internship offers a solution to this problem. The research will focus on data generated by single molecule sequencing technologies, with a special emphasis on the HellScope platform, which has a non‐uniform read length and a particular error model dominant with indels. Incorporating these features with a simultaneous mapping of multiple reads appears to be an NP hard problem. The research team will use specialized combinatorial optimization algorithms that provide provably approximate solutions for the worst case, to identify reads with overlapping alignments. The intern will monitor the convergence characteristics of the multi‐mapping calculations, and will default to reporting a subset of coordinates for reads with mapping multiplicity of more than a set of threshold.

Analysis of Decision Support System Metrics

Octothorpe Software is a decision sciences company that offers systems which organizations can use to improve their performance. This MITACS ACCELERATE internship project will be focused on developing new methods and metrics that organizations can use to improve their competitiveness. This will be achieved by developing the procedures, processes and the tools for displaying and understanding the key factors to improving the organizations success.

Aerosol Transport in Rooms with Displacement Ventilation

Ventilation design can influence the transport of infectious aerosols in buildings as well as the energy consumption of the building. Displacement ventilation (DV) holds the promise of reducing aerosol transmission and energy consumption. Stantec, a provider of professional design and consulting services in planning, engineering, architecture, surveying, and project management, has already investigated DV in hospital rooms, but that work did not consider the transient nature of aerosol sources (coughs and sneezes), or the physics of transportation and transformation of moist aerosols. This Mitacs-Accelerate internship will start with additional analysis of measurements gathered by the previous research on DV in hospital rooms, followed by field measurements in Greater Vancouver, in parallel with fundamental new experiments performed at UBC.

Integration of Technology to Support Nursing Practice: A Saskatchewan Initiative

This internship will focus on the potential of laptops and personal digital assistants (PDAs) to enhance the quality of work life for nurses through uptake of the technology to enhance access to information, and evidence-informed practice. The project is a partnership between College of Nursing, University of Saskatchewan and the All Nations’ Healing Hospital in Fort Qu’Appelle. Participation in this research project will allow the All Nations’ Healing Hospital to study the impact of introducing electronic nursing resources through the use of PDAs and laptops to assist the nursing staff in accessing timely and up to date policies and procedures, and best practice resources. The internship will also evaluate the impact of this project in supporting quality patient care and improving work life satisfaction for nurses. This research contributes to an identified research gap related to the applications and implications of ICT within the Saskatchewan health care workplace.

Neural Correlates of Branding Effects and Preference

Realtime Visualization and Sensor Fusion for an Action Sports Goggle Display

Water Pipeline Monitoring Using Wireless Sensor Networks

The intern will work closely with Pressure Pipe Inspection Company, a provider of critical water infrastructure management tools, to develop a new wireless data transmission system. The tasks will include 1) Rebuild Windows CE kernel for communication programming; 2) Data uploading program based on Winsock; 3) Data transmission software development between sites; 4) Design/Implement command function from the remote node. 5) Test the wireless data transmission system. 6) Improve the stability of the wireless system.

Tillage tool wear simulation using Discrete Element Method

Prepared samples of different metals with simple geometries will be worn in a lab‐based soil tillage operation. The conditions of the tillage operation will be maintained within defined limits (soil moisture content, compaction, and soil texture). At regular intervals of travel, the amount of material removed from the samples during the tillage operation will be quantified. The MITACS ACCELERATE intern will use dedicated software to simulate the wear process during the tillage operation and compare the simulation results with the experimental data. The comparative analysis will include the location and quantities of material removed during tillage. After the accuracy of the model is confirmed, it may then be used to predict the performance of other materials exposed to the same (or different) environment. The model will also be able to predict the response of the samples if the wear process was to continue beyond the duration of the experimental tests.

Reaching for a Better Understanding of the Portfolio Optimization Process

Fund managers must select a portfolio of assets from a group of securities so as to balance the conflicting goals of maximizing portfolio return and minimizing portfolio risk. There are many elegant approaches to this problem. All suffer from the same flaw – they all require estimates of the expected future risk and return of the constituent securities as well as of the correlations between the returns of pairs of these securities. Such estimates are difficult to obtain for two reasons: First, both companies and market conditions change over time making past history no indication of future performance. Second, the daily fluctuations in security prices tend to be much larger than any daily trend, making estimation of securities returns correlations very difficult. In this project, with Mapleridge Capital Corp, an alternative investment management company, the intern will investigate the impact of uncertainties in the parameters on the optimization process and develop robust methodologies for portfolio selection, using mathematical, statistical and computational techniques.

Quality Assessment of JPEG and JPEG2000 Compressed Medical Images

The goal of this project with Agfa Healthcare, a provider of diagnostic imaging and healthcare IT solutions, is to develop mathematical models for (i) assessment and (ii) improvement of compressed medical images. The rapidly increasing volume of data generated by new imaging modalities (eg CT scanner, MRI) necessitates the use of lossy compression techniques to decrease the cost of storage and improve the efficiency of transmission over networks. Increasing the degree of compression of an image, however leads to decreasing fidelity. One of the causes of this error is the introduction of artifacts by the compression method. Such distortions impede the ability of radiologists to make confident diagnoses from compressed medical images. The intern, in consultation with radiologists, will identify the important features that become degraded as the compression level is increased. In particular the research team shall study the JPEG and JPEG2000 compression schemes. The next step, of particular interest to Agfa, is to use this information to devise other, perhaps related, compression schemes that will not degrade these features at specific levels of compression. A second fundamental aspect of this work is to examine the use of “image quality measures” in the assessment of compressed medical images.

Image Rectification for 3D Stereoscopic Displays

3D, or Stereoscopic video technology has entered a new era. 3D stereoscopic effect has always relied on the viewers wearing special glasses to allow left and right image to be seen by each respective eye to create the stereoscopic effect of depth. Some industry players have started to investigate ways to create a 3D effect on the screen or display that does not the user to wear glasses. Spatial View Inc. is becoming a leader in researching, advancing, and developing stereoscopic displays that can be used in a variety of application areas from the computer desktop , to home theatre, to large screen advertisements to mobile phones and PDA’s. This internship will focus on one aspect of this research that deals with identifying points in left and right eye images to help automate the process of creating a perfect stereoscopic image.