Meet HPC Innovator Taghrid Samak
Everything leading up to the actual coding, figuring out how to make it work, is what Samak enjoys most
Taghrid Samak of Berkeley Lab’s Computational Research Division admits with a laugh that she wasn’t one of those kids who started programming on the home computer at age 10. And if she hadn’t followed her father’s advice, she might have ended up looking for political solutions to pressing problems, rather than working on computational approaches to scientific challenges.
One of the problems she is working on with the Department of Energy’s Joint Genome Institute (JGI) is a data mining method to automatically identify errors in genome assembly, replacing the current approach of manually inspecting the assembly. Genome sequencing is the process of determining the entire DNA sequence of an organism and plays an important role in genomic medicine, synthetic biology and other fields. DNA is extracted, then cloned in small fragments, sequenced and reassembled. One researcher compared the process to taking a pile of shredded newspaper and trying to put it together as a readable document — without knowing what it originally looked like.
Researchers at JGI have developed software to predict the likelihood of errors in the assembled sequences by comparing the assembled pairs with those from the original reads. When mismatches are identified, they can indicate errors in the assembly. For biologists to get the most benefit from the sequenced information, they need to be able to find and correct the errors. Samak’s work is aimed at being able to predict the type of error, as well as pinpoint the location in the assembly where possible errors occur. It’s difficult, as the data and results are quite “noisy,” and correct pairs near mismatched pairs can give a false positive.
But it’s exactly the kind of challenge Samak enjoys. As a schoolgirl growing up in Alexandria, Egypt, she loved math. “When the teacher gave out the problem sheet, I solved every problem, not just the assigned ones,” she said. Her father was a math teacher and she often spent time in his library.
As a teenager, she thought of studying political science, but her father advised her to pursue engineering, adding that she could always go into politics later. “He didn’t push me, but I’m really grateful he advised me to follow my passion for math,” she said.
Unlike high schools in the U.S. which offer a number of electives, Egyptian schools offered only set courses in science, math or liberal arts, and students had to choose one of the three. Samak studied math, then entered the engineering program at Alexandria University, where she was one of 70 students accepted for the computer science concentration. “After I got in, I took over the computer at home — a Windows 95 machine,” she said. “I was not one of those kids working on a computer when I was 10.”
For the first two years, she studied math, physics, mechanics and science, then took programming and numerical analysis classes. The approach suited her — she enjoys learning as much about the nature of a problem as she does finding a solution. “I really love to read other people’s work to learn more about the problem. What I enjoy most is everything leading up to the actual coding, figuring out how to make it work.”
But she hasn’t lost her early interest in politics and solving social problems. In March 2013, Samak was at MIT, meeting with other professionals from Egypt to take concrete steps to address at least some of the pressing issues in her home country that launched the Arab Spring. She chaired the 2013 EgyptNEGMA (Networking, Entrepreneurship, Growth, Mobilization, and Action) conference (http://egyptnegma.org/conference), where 10 proposals for advancing social development in Egypt were presented and the top three chosen. Programs to help people with disabilities, provide educational resources for school dropouts and an organization aimed at connecting charitable sustainability projects with potential funders were selected for further support and development. The theme of this year’s conference was “Empower the Crowd,” which Samak said is an apt description of the social revolution in Egypt.
“There was lots of energy at the conference, especially from the young people who are working on the ground in Egypt for social change,” Samak said. “It was really inspiring to see their enthusiasm. If we don’t do anything, nothing will happen. If we are able to help them, I think others will join in and things will get better.”
And it’s important for the projects to have strong ties to non-government organizations in Egypt, she said. “They trust their own, and it also gives them hope that they are getting support from Egyptians who are outside the country,” she said. “The most important thing we can give them is support. There are many ways to transfer our experience, and we are trying to find the best way to do it.”
Education and networking are central to Samak’s experience. Not only was her father a math teacher, but her mother taught biology. After earning her bachelor’s degree in computer science at Alexandria University, Samak came to the United States to pursue a doctorate in computer science at DePaul University in Chicago, where she worked as a teaching assistant, research assistant and then lecturer.
While attending the annual Grace Hopper Celebration of Women in Computing conference, she met Deb Agarwal, head of Berkeley Lab’s Advanced Computing for Science Department, as well as recruiter Jeff Todd. After finishing her Ph.D., she joined the lab as a post-doc and is now a research scientist. In addition to her work on data mining for genome assembly quality assessment, she is also helping with predictive modeling as part of DOE’s Carbon Capture Simulation Initiative (www.acceleratecarboncapture.org/drupal), modeling of energy consumption measurements for large-scale computing systems and “fingerprinting” HPC codes for cybersecurity analysis.
Since the March meeting at MIT, Samak has begun helping the Nebny Foundation (http://nebny.com/index.php), a non-profit organization in Egypt that is focusing on developing programs in education, health and entrepreneurship in disadvantaged neighborhoods. The U.S. branch of the Nebny Foundation is being incubated by TechWadi, a group aimed at building bridges for entrepreneurship between Silicon Valley and the Arab world.
Samak also maintains ties to Egypt and the Middle East through other outreach projects. For example, each summer she tutors students of her former professors in Alexandria, conducting weekly sessions over Skype. She even helped one of those students attend the SC12 conference in Salt Lake City.
For the past two years, she has also participated in TechWomen (www.techwomen.org), a U.S. Department of State initiative to bring women who are technical leaders from the Middle East and Africa for a month of mentoring and exchange of ideas with Silicon Valley companies and visits to Berkeley Lab. Samak has volunteered as a mentor for some of the participants during their visit to the U.S.
“With every student I talk to, I explain to them how computing helps science and improves our lives,” Samak said. “That usually has a big impact on them.”
Jon Bashor is a communications manager at Lawrence Berkeley National Laboratory. He may be reached at editor@ScientificComputing.com.