Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

The Author File

Shai Shen-Orr

Comparing single-cell trajectories with a new tool and what happens when aversion turns into love.

Shai Shen-Orr Credit: N. Shen-Orr

Biology is a comparative science, says Shai Shen-Orr, a computational biologist at the Technion–Israel Institute of Technology. When activated, cells might begin to express genes or differentiate, each in a slightly different way. “No two cells are created equal,” he says. Leveraging variation in gene expression in many single cells at one time point, labs can build cell trajectories. These trajectories are models of how single cells change over time. They are based on data that take much time and resources to generate, yet tools are lacking for trajectory comparison, he says. That's where cellAlign will come in handy, he says about his labs' new algorithm and software tool. “What cellAlign does is it allows you to compare trajectories, to compare between two trajectories, or different genes within a trajectory while maintaining the same high resolution that you spent so much effort to obtain,” he says.

Resolution is power when comparing single-cell trajectories. In a low-resolution photo, it's hard to tell an orange and tangerine apart except to see neither is an apple, says Shen-Orr. In this vein, cellAlign offers a way to compare trajectories at high resolution in terms of both local and global alignment. “The back-end math is the same as in sequence alignment,” he says. Just as with BLAST and sequence alignment, cellAlign applies dynamic programming to pick up differences between trajectories and patterns in common.

“Suddenly, I fell in love.”

The idea for cellAlign arose when Shen-Orr's team wanted to compare trajectories generated in the lab but didn't find tools they wanted. Around 40% of the lab does wet-lab work and the others take computational approaches, a blend of Shen-Orr's background.

Back in high school, Shen-Orr was averse to biology. His undergraduate focus at Technion was on machine learning and artificial intelligence. During a student exchange program at the University of Pennsylvania he took a database class and told the instructor Susan Davidson, co-director of the center for bioinformatics, that he urgently needed a part-time job because the IT job he had lined up had fallen through. She arranged for him to work for the center's director, Chris Overton, who has since passed away.

Shen-Orr was happy to put computing to work in a “humanitarian” context involving biomedical data and there was an allure of discovery unlike traditional IT projects. “Suddenly, I fell in love,” he says. By the end of his stay at UPenn, he was keen on learning more biology. He finished his Technion program, obtained a master's degree in bioinformatics at Weizmann Institute where he worked on E. coli data and then he completed his PhD in biology at Harvard University with Craig Hunter where he studied C. elegans. Human biology drew him in. “I wanted to be closer to medicine,” he says. As “a child of the genomics era,” it had become possible for him to make this leap. During his PhD studies, his burgeoning interest in immunology was reinforced by personal circumstance: his father faced and survived lymphoma. Shen-Orr was a postdoctoral fellow with immunologist Mark Davis at Stanford University and then joined the faculty at the Technion.

“Shai's defining attribute is his infectious enthusiasm and optimism,” says systems immunologist John Tsang, who co-directs the Center for Human Immunology at the National Institutes of Health. He was drawn to Shen-Orr's energy, geekiness and desire to affect human health by entering the field of computational human immunology after his PhD. The two men met in graduate school and have since collaborated. Immunology, with its blend of molecular, cellular and informational complexity along with its real-world impact, fits Shen-Orr perfectly, he says.

In 2016, Shen-Orr co-founded a company, CytoReason, to build a machine-learning model of the human immune system. For example, the team mines the published literature and offers customers a way to look at their data in the context of what has been published.

Besides the lab and his company, Shen-Orr covets time with his wife and two young daughters, whom he wants to empower “to do anything they set their mind to.” Women are the majority in his lab, which has “just evolved, it's not anything intentional,” he says. He assumes word has spread that the lab is welcoming to women scientists. “Certainly we have a lot of mothers in the lab, and I fully respect that,” he says. He offers mothers flexibility as needed and says there have been nine births since the lab was launched. A lab is “about thinking about science and dreaming science and being productive,” he says. “Family is also important.”

References

  1. 1

    Alpert, A., Moore, L.S., Dubovik, T. & Shen-Orr, S.S. Alignment of single-cell trajectories to compare cellular expression dynamics. Nat. Methods 15, 267–270 (2018).

    CAS  Article  Google Scholar 

Download references

Authors

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Marx, V. Shai Shen-Orr. Nat Methods 15, 231 (2018). https://doi.org/10.1038/nmeth.4659

Download citation

Search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing