MS Thesis Defense

Finding associations among SNPs for
prostate cancer using collaborative filtering

Rohit Kugaonkar

9:00am Wed. 18 July 2012, Room ITE 325b

Prostate cancer is the second leading cause of cancer related deaths among men. Because of the slow growing nature of prostate cancer, sometimes surgical treatment is not required for less aggressive cancers. Recent debates over prostate-specific antigen (PSA) screening have drawn new attention to prostate cancer. Due to the complicated nature of prostate cancer, studying the entire genome is essential to find genomic traits. Due to the high cost of studying all Single Nucleotide Polymorphisms (SNPs), it is essential to find tag SNPs which can represent other SNPs. Earlier methods to find tag SNPs using associations between SNPs either use SNP's location information or are based on data of very few SNP markers in each sample. Our study is based on 2300 samples with 550,000 SNPs each. We have not used SNP location information or any predefined standard cut-offs to find tag SNPs. Our approach is based on using collaborative filtering methods to find pair wise associations among SNPs and thus list top-N tag SNPs. We have found 25 tag SNPs which have highest similarities to other SNPs. In addition we found 16 more SNPs which have high correlation with the known high risk SNPs that are associated with prostate cancer. We used some of these newly found SNPs with 5 different classification algorithms and observed some improvement in prediction accuracy over using the original known high risk SNPs. The classifier can be used in a decision to perform further testing in case of a "yes" answer by the classifier.

Committee: Drs. Yelena Yesha (chair), Anupam Joshi, Aryya Gangopadhyay and Micheal Grasso.

Professor Marie desJarins (PI) along with co-principal investigators Professor Penny Rheingans and Dr. Susan Martin received a research award worth $200,000 from the National Science Foundation's program on Computing Education for 21st Century. Their 18 month research project  will gather data about the status of Computer Science education in Maryland high schools and build relationships among high school teachers, community college and university faculty, and state education administrators to facilitate and increase state-level support for lasting improvements to computing education.

Despite the overall success of the K-12 education system in Maryland, opportunities to study computer science vary tremendously among the 24 school systems and approximately 200 high schools in the state. This disparity can be attributed to several factors, including the lack of a state-mandated computer science high school graduation requirement, the fact that there is no state-required teacher certification in the discipline, the absence of a standardized computer science curriculum, and barriers to entry for girls and underrepresented minorities.

The effort has two immediate objectives: (1) performing an assessment of the current state of high school computer science in each of the 24 Maryland school systems and (2) increasing knowledge about national issues associated with computer science education among high school and state administrators in Maryland through state-wide summit meetings for teachers, administrators, and higher education faculty. The long term goal is to develop curriculum and teacher development programs that will improve the quality, breadth, and student diversity of computer science education in Maryland.

This week the fencing around UMBC’s new Performing Arts and Humanities Building came down, making the area behind the ITE and ECS buldings more attractive and access to parking more convenient. The PAHB building has been turned over to UMBC and furniture will be delivered next week. It will be open later this summer in time for the Fall 2012 semester. Construction continues on the second phase of the building.

The PAHB will be home to the Department of Ancient Studies, the Department of Dance, the Department of English, the Department of Music, the Department of Philosophy, the Department of Theatre, the James T. and Virginia M. Dresher Center for the Humanities, the Humanities Scholars Program, and the Linehan Artist Scholars Program.

Business Intelligence Internship (6/25/2012)

Optimal Solutions Group LLC (“Optimal”) has an opening for a Business Intelligence (BI) Intern at its offices at the University of Maryland’s Research Park at M-Square. This BI intern will assist research and IT personnel on multiple projects. Optimal differentiates itself from other public policy research and technical assistance firms by its quantitative expertise and innovative approaches. Founded in 2000, Optimal has more than 30 active federal contracts and a multidisciplinary staff of over fifty employees with expertise in health, education, workforce development, and information technology.

The qualified BI intern will perform the following duties:

• Interact with research and IT staff to gather Business Intelligence and reporting requirements , help prioritize the requirements, create technical specifications
• Design, code, test, and document all new or modified BI systems, applications, and programs
• Assist in the design of databases and data warehouses to ensure interoperability with BI solutions
• Perform routine program and system modifications, performance tuning of applications as necessary, problem solving and debugging
• Support the on boarding and training of Business Users and Business Analysts
• Design and deliver end-user training and training materials; provide technical support as necessary
• Adhere to good systems development methodology
• Conduct research and make recommendations on BI products, services, and standards in support of procurement and development efforts
• Participates in special projects as required

Required Specialized Knowledge and Skills:

• Experience with SQL server databases and database concepts
• Good understanding of Object Oriented programming concepts and SDLC processes
• Experience in Spotfire, MicroStrategy, Cognos, or other BI Tools
• Good understanding of Data warehousing and data mining concepts
• Has knowledge of commonly-used concepts, practices, and procedures within the BI reporting field.
• Strong Communication skills (Verbal and written)
• Must have strong analytical, troubleshooting, and problem solving abilities.
• Must have the ability to work independently with minimal supervision.
• Must have the ability to work under pressure and adhere to deadlines.

Education and Experience:

• Must be actively pursuing a B.S. or Master’s degree. Majors in Computer Science, Management Information Systems, computer information Systems, or Computer Engineering preferred.

Work Hours and Compensation:

• Part-time during the school year and full-time during the summer. Compensation is commensurate with years of education, training, and pertinent experience.

To Apply:

• Submit cover letter and resume to Sorry, you need javascript to view this email address. , reference “BI Intern” in subject line.

Ph.D. Dissertation Defense

Relay Node Placement for
Federating Segmented Wireless Sensor Networks

Fatih Senel

2:00pm Tuesday, 10 July 2012, ITE 325b

Recent years have witnessed a growing interest in the applications of Wireless Sensor Networks (WSNs). Most notable among these applications are those operating in hostile environments space exploration, border protection, combat field reconnaissance, and search and rescue. Due to the harsh surroundings, WSNs may suffer from a large scale damage that causes many nodes to fail simultaneously and the network to get partitioned into multiple disjoint segments and its services become very limited. In such a case, restoring the network connectivity is very important in order to avoid negative effects on the applications. Linking disjoint segments may not be feasible through coordinated repositioning of some set of nodes as the scope of the damage is so wide that cannot be determined. One of the viable solutions for federating damaged WSNs is to deploy additional resources, i.e. relay nodes, to form inter-segment multi-hop paths.

In this dissertation, we tackle technical challenges related to the federation of segmented WSNs. We present a set of effective techniques that for repairing the damaged WSN using the least number of relay nodes (RNs) as well as maintaining some desirable topology features such as robustness against failures, network coverage and balanced traffic load. The correctness and time-complexity of all proposed approaches are analyzed and their performance is validated through extensive simulation experiments.

Committee: Drs. Mohamed Younis (Chair), Charles Nicholas, Samuel Lomonaco, Tim Oates, Kemal Akkaya and Waleed Youssef

MS Thesis Defense

Integrating Domain Knowledge in Supervised Machine Learning
to Assess the Risk of Breast Cancer Using Genomic Data

Aniket Bochare

9:00am Friday 29 June 2012, ITE 325b

Breast cancer is the most common form of cancer in women. Breast cancer comprises 22.9% of the invasive cancers in women and 16% of all the female cancers. Currently, treatment decisions are based primarily on clinical parameters, with little use of genomic data. Our study takes into consideration the data of postmenopausal women of European descent and their single nucleotide polymorphism (SNP) information to assess the risk of developing breast cancer. We used various supervised machine learning and data mining techniques to generate a model for predicting risk of breast cancer using only genomic data.

In this research we propose an approach to select the nine best SNPs using various feature selection algorithms to improve binary classification accuracy and validate our results with the existing literature. The machine learning model generated without the domain knowledge yields poor prediction results. After the addition of the domain knowledge of the 11 SNPs into the original training set we performed classification using the best features obtained by feature selection techniques. The machine learning model generated using both the domain knowledge and the feature selection techniques performed much better compared to the naive approach of classification.

Committee: Drs. Yelena Yesha (chair), Anupam Joshi, Aryya Gangopadhyay and Micheal Grasso

It's got a bowling alley, a piano in the lobby, a T-Rex skeleton, and eighteen cafeterias. It might sound like a theme-park, but it's the Googleplex, a.k.a. Google headquarters, located in Mountain View, California.

This extremely enlightened workplace, which features a slew of employee benefits like a free shuttle to work, laundry services, and even haircuts, might sound like an un-productive paradise, but the benefits are actually put in place to offset all the hard work and dedication of the employees.

So says an article in the Summer issue of UMBC Magazine, where Computer Science alumni Jeffrey Burgan '83 (pictured), Sandor Dornbush '06 (pictured), Jeetendra M. Soneja '04, and Robert Banz '95 talk about what it's like to work for Google.

Check out the article here.

Photo Courtesy Gabriela Hasbun

Photo Courtesy UMBC Magazine

In its latest issue, dubbed "Tech Trekkers", UMBC Magazine features Computer Science alumnus Harry Chen, one of the key players in the development of the iPhone 4S's digital assisant, Siri.

While you may have heard about Harry through an article we posted last year, the UMBC Magazine article goes deeper into Harry's story as it  follows him around the Apple campus in Cupertino, California, where he works as a software engineer.

Weaving the inspiring tale of Harry's path to success–including his spontaneous decision to pack up everything and move to the West coast in 2008–the article paints Harry as a tech wunderkind who credits at least some of his success to the environment of discovery at UMBC.

It talks about his graduate research on smart rooms ("Imagine a room that answers your questions, or anticipates your needs and then fulfills them," goes the article) and features quotes from CSEE Professors Dr. Tim Finin and Dr. Anupam Joshi, who served as his advisors during his nearly ten years at UMBC, during which time he received a B.S., M.S., and Ph.D. in Computer Science.

Want to be inspired? Check out the article here.

Originally from Almora, India, Yatish came to UMBC in 2009 to pursue his Master's in Computer Engineering. When he's not studying, Yatish likes to run, play disc golf, and read thriller novels.

About Yatish…

When did you know you wanted to study Computer Engineering?

I pursued a Bio/Math track in high school and wanted to be a doctor more than an engineer, but by 12^th grade, all the dissections in Biology turned me off and finalized my choice of Engineering.

Are you part of any labs, clubs or organizations on campus?

Yes I am part of the ESNET Lab run by Dr. Younis.

What are you researching for your thesis?

My research area is Wireless Sensor Networks, specifically algorithms and approaches to recover a network from multi node failures efficiently using distributed approaches.

Have you had any internships? 

I have interned in Samsung Electronics in 2010, working on managing Visual Display Projects related to the Hotel TV lineup.

What are your plans after graduation?

I plan to go back and work for a few years before returning to academia.

What is your dream job?

Setting up my own technical consulting firm.

On UMBC and CSEE…

What was your first impression of UMBC?

I liked the fact that everything was enclosed in a loop so it was easy to get around, and I was surprised to find that we had a nationally ranked chess team and no football team!

What was your first Computer Engineering Course?

My first Computer Engineering course at UMBC was CMSC611: Advanced Computer Architecture taught by Dr. Younis. It is one of the most fundamental classes and was one of my most favorite and enjoyable class to date.

What was your favorite Computer Engineering Course?

I would recommend CMSC611: Advanced Computer Architecture, and CMPE640: VLSI design to anyone in the CSEE program to build a solid foundation. People that enjoy playing with numbers should definitely take CMPE 645: Computer Arithmetic Algorithms.

What do you like about the Computer Science and Electrical Engineering (CSEE) Department?

I love the fact that professors and staff are easily accessible and you can stop by to ask questions or just chat.

How would you describe the professors in the department?

They are friendly and willing to help you out outside class.

What advice would you give to incoming graduate students?

Don’t forget to have fun, despite the workload.

This video is a short documentary about the LEGO Turing Machine built by Jeroen van den Bos and Davy Landman at Centrum Wiskunde & Informatica in Amsterdam.  They built the device for CWI's exposition "Turings Erfenis" in honor of Alan Turings one hundredth birthday this year and to demonstrate how simple a "universal computing device" can be.

Alan Turing was a brilliant mathematician who helped define the theoretical model of the computer as we know it today. He also made many other significant contributions to cryptology, artificial intelligence and computer sceince. He was a visionary, one of the few people of his time who recognized the role the computer would play for humanity.

The Turing Machine was defined in 1936 as an abstract model for a simple device that manipulates symbols on a strip of tape according to a table of rules. Turing Machines are not intended to be practical, but can be used to simulate the logic of any computer algorithm and to define the limits of what can be computed by real computers. You can learn more about the abstraction and why it is important by taking CMSC 451, Automata Theory and Formal Languages.

There have been many physical models built in the past, but this one is interesting because it was built with a single Lego Mindstorms NXT set. One unfortunate limitation of this Turing Machine is that it has a finite 'tape'. See the Lego Turing Machine site for more information.