In the United States, the Next Generation Science Standards advocate for the integration of computational thinking (CT) as a science and engineering practice. Additionally, there is agreement among some educational researchers that increasing opportunities for engaging in computational thinking can lend authenticity to classroom activities. This can be done through introducing CT principles, such as algorithms, abstractions, and automations, or through examining the tools used to conduct modern science, emphasizing CT in problem solving. This cross-case analysis of nine high school biology teachers in the mid-Atlantic region of the United States documents how they integrated CT into their curricula following a year-long professional development (PD). The focus of the PD emphasized data practices in the science teachers' lessons, using Weintrop et al.'s definition of data practices. These are: (a) creation (generating data), (b) collection (gathering data), (c) manipulation (cleaning and organizing data), (d) visualization (graphically representing data), and (e) analysis (interpreting data). Additionally, within each data practice, teachers were asked to integrate at least one of five CT practices: (a) decomposition (breaking a complex problem into smaller parts), (b) pattern-recognition (identifying recurring similarities in data practices), (c) algorithms (the creation and use of formulas to predict an output given a specific input), (d) abstraction (eliminating detail in order to generalize or see the "big picture"), and (e) automation (using computational tools to carry out specific procedures). Although the biology teachers integrated all CT practices across their lessons, they found it easier to integrate decomposition and pattern recognition while finding it more difficult to integrate abstraction, algorithmic thinking, and automation. Implications for designing professional development experiences are discussed.