[10][11] One of his papers on the interval satisfiability problem was later applied to the study of DNA physical mapping;[12] this marked his introduction to the field of computational biology.

Shamir used his expertise in graph theory to develop clustering algorithms for analyzing gene expression problems.

[13] His CAST algorithm, with Zohar Yakhini and Amir Ben-Dor was published in 1999[14] and drew a lot of attention from the bioinformatics community; the techniques described in the paper became popular for analyzing genomic data.

Shamir broadened his research to include additional aspects of bioinformatics, such as analysis of biological networks,[17][18] genome rearrangements,[19] sequence motif finding,[20][21] and transcriptional regulation.

[22][23] Many tools developed in his laboratory are available as a part of the EXPANDER suite,[6] which provides an integrated environment for analyzing high-throughput biological data.

Shamir's current research focuses on integrative analysis of heterogeneous high-throughput bio-medical data, genome rearrangements in cancer, and gene regulation.

Shamir was on the founding steering committee of the RECOMB meeting,[25] the premier theoretical conference in bioinformatics, and served on it for thirteen years.

Among his students in academia are Haim Kaplan, Dekel Tsur, Dalit Naor, Itsik Pe'er, Roded Sharan, Amos Tanay, Adi Akavia, Reut Shalgi, Rani Elkon, Rotem Sorek, Irit Gat-Viks, Michal Ziv-Ukelson, Igor Ulitsky, Mukul Bansal, Meirav Zehavi, Yaron Orenstein, Adi Maron-Katz and Lianrong Pu.