Deriving actionable insights from data requires expertise in both data science as well as the specific application domain. This need for domain-specific knowledge often necessitates engaging an interdisciplinary team rather than an individual for most realistic data science problems in domains such as finance, biology, or drug discovery. This, in turn, requires effective collaboration between team members. This paper seeks to understand common themes in how such multi-disciplinary teams communicate to accomplish their analytical goals. We conduct an interview study with 15 professional data scientists working in small to large organizations in fields ranging from bioinformatics to accounting. Communication between individuals in these teams depends on their team structure and expertise. Data scientists specifically adapt their tools to communicate with team members with different types of domain knowledge. We discuss the strengths and weaknesses of these approaches for supporting communication in multi-disciplinary environments.