BioEngineering

Click on the title for full description of SURE 2017 projects in BioEngineering.

BIO-001:  Biocomputation with ‘Smart’ Biological Agents
Professor: Dan Nicolau
E-mail: dan [dot] nicolau [at] mcgill [dot] ca
Telephone: 514-718-8261
Website 

Research Area: Bioengineering


Description:
Many mathematical and real-life problems cannot, or are very difficult to be solved by the present computers which process the information sequentially and with extreme precision. Among these problems one can mention travel and production scheduling, class time tables, and cryptography. Despite this difficulty, these problems are solved easily by individual biological agents, from microorganisms to humans, who do not process the information sequentially, but in parallel, and who trade precision for heuristic decision making. Alternatively, some mathematical and real-life problems that cannot be solved by the present computers are also difficult to solve by individuals, due to the limited capacity of an individual to process the information in parallel, but can be solved heuristically by groups of individuals operating together either explicitly or tacitly. Among these problems one can mention behaviour of groups in panic situations, solving complex traffic problems, hierarchical self-organisation of groups in conflictual situations. To this end, the project aims to assess the individual and collective ‘computational power’ of individual biological agents in optimally partitioning the available space and taking optimal decisions. The possible applications range from medical to new algorithms and computer paradigms. The project involves either experiments, such as observing the ‘intelligent’ behaviour of microorganisms facing space confinement via their movement in microfabricated networks; or the modelling and simulation of their behaviour; or a combination of both. The ‘smart’ biological agent of choice is a fungus, which has been demonstrated as using intelligent algorithms for searching labyrinths.

Tasks:
The project can be approached, depending on the student’s strengths, either from an experimental, or a simulation perspective. Experimental tasks comprise the fabrication of simple microfluidics structures; growth of microorganisms in microfluidics structures; observation and recording of microorganisms behavior. Simulation tasks comprise the translation of microorganisms behavior in logic rules and simple algorithms; and the simulation of microorganisms behavior in complex structures.

Deliverables:
Report on the optimality of microorganisms behavior. In both cases one conference paper is expected at the end of the project.

Number of positions: 2
Academic Level: Year 3

BIO-002: Information Storage on Molecular Surfaces of Biomolecules
Professor: Dan Nicolau
E-mail: dan [dot] nicolau [at] mcgill [dot] ca
Telephone: 514-718-8261
Website

Research Area: Bioengineering


Description:
The spatial recognition of objects, from airplanes to human faces, is of ever-increasing interest in the present interconnected and crowded world. While this problem is tackled by humans by a myriad of image analysis and recognition algorithms implemented in dedicated software, a similar problem is seamlessly solved in Nature by the ‘image recognition’ between biomolecules – the cornerstone of all biological processes. However, and despite their theoretical similarity, presently only separate, specialised programs are used for image recognition for the macro-world, e.g., biometrics, and nano-world, e.g., drug discovery. The project will involve the use of existing in-house developed software for building images of biomolecules, followed by the development of an interface between structural databases, image building for the bio-objects present in these databases, and the archiving, classification and access to a database of molecular images.

Tasks:
Upgrade of the existing simulation procedure for molecular surfaces; running simulation for a small set of proteins; search for commonality of characteristics between the mapped molecular surfaces.

Deliverables:
Update the existing Biomolecular Adsorption Database (BAD); report regarding the property distribution on molecular surface. One conference paper is expected at the end of the project.

Number of positions: 1
Academic Level: Year 3

BIO-003: Investigation of dynamic functional connectivity using simulated and experimental multimodal neuroimaging data
Professor: Georgios Mitsis
E-mail: georgios [dot] mitsis [at] mcgill [dot] ca
Telephone: 514-398-4344
Website

Research Area:Signal processing, functional neuroimaging


Description:
The exceptional capacity of the brain to process complex stimuli arises largely from the presence of intricate interactions between different regions. Therefore, understanding connectivity holds one of the major keys for understanding brain function in health and disease. More recently, there has been much interest in dynamic functional connectivity, i.e. how connectivity patterns vary over time. In this context, the main objective of the present project is to use advanced signal and image processing to better understand the nature of dynamic functional connectivity using both simulated and experimental data. Specifically, we will develop forward computational models to generate realistic simulated data as well as analyze multimodal neuroimaging data (simultaneous EEG-fMRI and MEG) during resting-state conditions and motor task execution, collected at McGill’s Brain Imaging Center. We will specifically investigate the possible source of time-varying functional connectivity patterns, such as physiological factors (i.e. fluctuations in physiological signals such as heart rate, respiration and arterial CO2) and electrophysiological signatures, such as power in different frequency band of the EEG signal. To achieve this we will use nonstationary signal processing methods such as wavelets and time-varying multivariate autoregressive models that our lab has developed. Multimodal imaging methods are very promising to better elucidate the exploiting the excellent time resolution of EEG, MEG and the excellent spatial resolution of fMRI. Further validation of the motor connectivity measures to be identified will yield a set of robust and sensitive biomarkers of age-related motor decline, which may ultimately guide personalized treatment strategies using exercise or stimulation protocols (e.g. transcranial direct current stimulation).

Tasks:
The first student will focus on generating and analyzing realistic simulated data, based on prior work conducted in our lab. The focus of this student's work will be to generate several plausible scenarios may give rise to time-varying connectivity patterns, such as physiological fluctuations, and quantify their effects. The second student will focus on analyzing experimental data, particularly simultaneous EEG-fMRI data using advanced methods with an aim to better understand how signals recorded with different modalities may be used to quantify dynamic connectivity patterns, e.g. how does EEG signal power in different frequency bands affect the slow fluctuations observed with the fMRI BOLD signal?

Deliverables:
Student 1. Deliverable 1: Matlab toolbox implementing computational models for simulated data generation. Deliverable 2: Technical report. Student 2. Deliverable 1: Processing pipeline for analyzing the multimodal experimental data. Deliverable 2: Technical report.

Number of positions: 2
Academic Level: Year 3

BIO-004: Computational structural biology: Evolutionary design principles of proteins
Professor: Yu Xia
E-mail: brandon [dot] xia [at] mcgill [dot] ca
Telephone: 514-398-5026
Website

Research Area:Bioinformatics, Computational Biology


Description:
Proteins are evolved molecular machines capable of self-assembly and reliable functioning in fluctuating environments. Understanding the physical and evolutionary principles underlying these remarkable properties of proteins is a central challenge in biomolecular engineering. This project will focus on computer modeling of protein structure and evolution. Homology modeling will be used to construct three-dimensional structural models of various proteins. Next, structural, biophysical and evolutionary properties of these proteins will be investigated, with the aim to understand how biophysical properties of proteins affect their evolutionary properties at the residue level. The focus will be on soluble and membrane proteins.

Tasks:
Literature review; becoming familiar with existing publicly-available datasets on protein sequence and structure; becoming familiar with computational tools on modeling protein structure and evolution; computer programming.

Deliverables:
A final report summarizing the findings.

Number of positions: 2
Academic Level: Year 3

BIO-005: Computational systems biology: Design principles of protein networks
Professor: Yu Xia
E-mail: brandon [dot] xia [at] mcgill [dot] ca
Telephone: 514-398-5026
Website

Research Area:Bioinformatics, Computational Biology


Description:
The cell is the fundamental unit of life, yet the inner workings of the cell are far more complex than we ever imagined. Without a good model of the cell, it is difficult to develop new drugs to repair diseased cells, or build new cells to produce much-needed chemicals and materials. A key step towards building a working model of the cell is to map the complex network of interactions between thousands of tiny molecular machines in the cell called proteins. This project will focus on computer modeling of protein networks. Various publicly-available datasets on protein networks will be integrated and visualized. The resulting integrated protein networks will then be annotated with evolutionary and disease properties, with the aim to understand how protein networks evolve, and how disruptions in protein networks lead to disease. The focus will be on protein networks in yeast and human.

Tasks:
Literature review; becoming familiar with publicly-available datasets on protein networks; becoming familiar with existing computational tools on modeling protein networks; computer programming.

Deliverables:
A final report summarizing the findings.

Number of positions: 2
Academic Level: Year 3

BIO-006: Optical tweezers for single-molecule studies of motor proteins and cell mechanics
Professor: Adam Hendricks
E-mail: adam [dot] hendricks [at] mcgill [dot] ca
Telephone: 514-893-2343
Website

Research Area:Bioengineering, Cell mechanics, Intracellular transport


Description:
Optical tweezers (or optical traps) use a tightly-focused laser beam to exert forces on micron-sized refractive objects. By attaching motor proteins to small latex beads, we can measure the forces exerted by single molecules. Our lab has also developed techniques to measure the forces exerted by motor proteins and characterize the viscoelastic environment in living cells. Here, we will modify our current optical trapping systems to add two important capabilities. First, we will develop a force-feedback optical trap that allows us to exert constant forces on motor proteins as they move along cytoskeletal filaments. The force is measured by collecting the light that passes through the bead onto a quadrant photodiode, and the the position of the trap is controlled through an acousto-optic deflector. Second, we will develop the ability to simultaneously manipulate several beads by rapidly switching the position of the optical trap such that multiple time-shared optical traps are formed. Multiple traps will be used to measure the mechanical response of the cell over several length-scales.

Tasks:
Student 1: (1) Develop optical tweezers capable of manipulating single molecules and measuring their nanometer-sized displacements and pN-level forces. (2) Program a simple feedback controller to maintain constant forces. Student 2: (1) Develop software to control the acousto-optic deflector to form multiple, time-shared optical traps. (2) Use the optical tweezers to examine the viscoelastic properties of the cellular environment.

Deliverables:
Student 1: (1) System capable of applying constant forces using a feedback controller. (2) Recordings of the movement of single motor proteins under constant forces. Student 2: (1) Control system to create multiple time-shared optical traps. (2) Measurements of intracellular mechanics over multiple length scales.

Number of positions: 2
Academic Level: No Preference

BIO-007: Regulation of motor proteins in intracellular transport and cell division
Professor: Adam Hendricks
E-mail: adam [dot] hendricks [at] mcgill [dot] ca
Telephone: 514-893-2343
Website

Research Area:Bioengineering, Motor proteins and the cytoskeleton, Single-molecule biophysics


Description:
The motor proteins kinesin and dynein move along microtubules to transport cargoes and organize microtubules in the cell. Our goal is to understand how multiple motor proteins operate in teams, and how they are regulated to perform complex functions like cell division and directed transport. Through extending single-molecule techniques to native organelles and living cells, we have developed advanced microscopy tools to measure the regulation, motility, and forces exerted by motor proteins with unprecedented resolution, and to manipulate the system by applying external forces to the cargoes through optical tweezers and controlling motor activity using optogenetics. We will image and manipulate ensembles of kinesin and dynein as they transport native cargoes in reconstituted systems and living cells to understand how kinesin and dynein motors interact, how they are controlled to direct intracellular trafficking and cell division, and how motor proteins are misregulated in neurodegenerative disease and cancer.

Tasks:
Student 1: (1)Student 1: Express and purify proteins and organelles. Perform single-molecule in vitro motility assays. Analyze images. Student 2: Develop micro patterned microtubule arrays to reconstitute spindle assembly in vitro.

Deliverables:
Student 1: (1)Student 1: Analysis of the role of the scaffolding molecule huntingtin in regulating kinesin and dynein motility. Student 2: Protocols to micro pattern spindle-like microtubule arrays. Analysis of kinesin-5 motility and crosslinking on reconstituted microtubule arrays.

Number of positions: 2
Academic Level: No Preference

BIO-008: Bioprocess supervision through data acquisition, integration and control of Critical Process Parameters
Professor: Amine Kamen
E-mail: amine [dot] kamen [at] mcgill [dot] ca
Telephone: 514-398-5775
Website

Research Area:Bioprocess Intensification for Production of Viral Vaccines


Description:
Bioprocess development involves upstream, and downstream processing; and analytical process technology optimization and integration. A significant amount of monitoring and control data are generated at all steps of the bioprocess. These data are derived from different types of equipment: bioreactors, purification units and analytical units. The aim of the project is to design data bases and streamline data acquisition and integration in historical display for supervision and control of advanced bioprocesses.

Tasks:
Develop good knowledge of bioprocess steps - Access communication protocols from equipment manufacturers - Design a network for streamlining process data acquisition - Develop drivers for communication data base/equipment units

Deliverables:
Data base integrating process operation - Drivers for equipment unit - User manual

Number of positions: 1
Academic Level: Year 3

BIO-009: Plasmonics-based biosensors
Professor: Sebastian Wachsmann-Hogiu
E-mail: sebastian [dot] wachsmannhogiu [at] mcgill [dot] ca
Telephone: 425-287-4211

Research Area:"Biosensors Point of care technologies"


Description:
The project will focus on the development of biosensing technologies based on optical spectroscopy and microscopy in combination with plasmonics and microfluidics. The goal will be the optimization of nanostructures for improved reproducibility, sensitivity, and specificity. The ability of these structures to work as analytical devices will be explored.

Tasks:
Student 1 - Learn to prepare plasmonic nanoparticles and substrates.Perform optical measurements (spectral, etc.). Student 2 - Prepare experimental set-up. Collect and analyze data.

Deliverables:
Student 1 - Metallic nanostructures of different size and composition. Spectra of thiolated molecules attached to nanostructures. Student 2 - Experimental setup for optical measurements. Prepare microfluidics device. Statistical analysis of spectra involving Principal Component Analysis and regression methods.

Number of positions: 2
Academic Level: No Preference

BIO-010: Mathematical modeling of tumor growth and the effects of therapy with 3D cell cultures *** posted January 24th, 2017
Professor: Georgios Mitsis
E-mail: georgios [dot] mitsis [at] mcgill [dot] ca
Telephone: 514-398-4344

Research Area:Computational oncology


Description:
Despite the unquestionable progress in basic cancer research, which has included the use of computational approaches, the widespread application of the latter to cancer therapy design in a clinical setting is still elusive. However, quantitative approaches can significantly contribute towards better understanding the underlying biological mechanisms as well as the long-held goal of designing patient-specific therapeutic strategies. Specifically, constructing mathematical models that can reliably predict tumor growth and its response to therapy according to a patient’s individual characteristics can be used to achieve the latter goal. In this context, the present project aims at building spatiotemporal mathematical models describing tumor growth and the effects of therapy and validating these models using state-of-the-art 3D bioprinting (cell culture) methods. Specifically, we will use data generated from 3D bioprinting techniques developed in the research lab of Prof. M. Kinsella (Bioengineering), which are able produce in vitro models of tumor tissue capable of mimicking the mechanical, pathophysiological, and cellular heterogeneous state found in native tumors. The parameters of the developed models will be fine-tuned using the experimental data and the effect of varying several experimental parameters (e.g. initial position and concentrations of tumor cells and fibroblasts in the cultures) will be quantitatively assessed.

Tasks:
The student will first conduct a literature survey related to spatiotemporal models describing tumor growth and therapy effects. Based on that, he/she will select the model structure that is most suitable for the available experimental data (3D cell cultures). The selected model structures will be subsequently fitted to the data in order to provide better quantitative understanding of the underlying mechanisms and assess the effect of different experimental design parameters related to the 3D cultures.

Deliverables:
Deliverable 1: Matlab toolbox implementing computational models for tumor growth and therapy effects. Deliverable 2: Technical report.

Number of positions: 1
Academic Level: Year 3

Follow us