Background: The disease-causative variant remains unidentified in approximately 0.5% to 2% of hemophilia B patients using conventional genetic investigations, and F9 deep intronic variations could be responsible for these phenotypes.Objectives: This study aimed to characterize deep intronic variants in hemophilia B patients for whom genetic investigations failed.Methods: We performed whole F9 sequencing in 17 genetically unsolved hemophilia B patients. The pathogenic impact of the candidate variants identified was studied using both in silico analysis (MaxEntScan and spliceAI) and minigene assay.Results: In total, 9 candidate variants were identified in 15 patients; 7 were deep intronic substitutions and 2 corresponded to insertions of mobile elements. The most frequent variants found were c.278-1806A>C and the association of c.278-1244A>G and c.392-864T>G, identified in 4 and 6 unrelated individuals, respectively. In silico analysis predicted splicing impact for 4 substitutions (c.278-1806A>C, c.392-864T>G, c.724-2385G>T, c.723+4297T>A). Minigene assay showed a deleterious splicing impact for these 4 substitutions and also for the c.278-1786_278-1785insLINE. In the end, 5 variants were classified as likely pathogenic using the American College of Medical Genetics and Genomics guidelines, and 4 as of unknown significance. Thus, the hemophilia B-causing variant was identified in 13/17 (76%) families.Conclusion: We elucidated the causing defect in three-quarters of the families included in this study, and we reported new F9 deep intronic variants that can cause hemophilia B.