We divided our dataset (Table 2) into two datasets representing RISs identified in lymphoid (65) and myeloid (83) leukemias. A total of 23 RISs were excluded, as they were identified in animals with no clear-cut diagnosis. The 2 individual datasets were subjected to pathway analysis (Table 2). P values reaching significance (< .05) are indicated in bold. Pathways fulfilling the following two criterias are depicted: (1) must have 3 proviral insertions in 3 or more genes in a least one of the 2 groups; (2) must display selectivity (ie, only one experimental group can reach significance; P < .05). Pathways are sorted based on significance. RIS affected in more than one tumor are depicted in bold, and the number of integrations in a given disease subtype is depicted in superscript.