Referee 2's Review:

Lozier et al describe a histological study of the livers of two mouse models, a Jagged1/Notch2 compound heterozygote and a Notch2 conditional homozygous knockout. The question the authors had in mind was to establish whether the histological liver defects of bile duct paucity resulted from defects in differentiation or defects in morphogenesis of the ductal plate. This is an interesting question and the genetic tools are perfectly poised to answer this. Furthermore, the findings for the Notch2 conditional are novel and very interesting and the evidence for necrosis and growth retardation is compelling.

The data are presented as a series of clear histological photographs and a table of blood analyses. The photographs and methods section do not indicate the numbers of animals examined nor the numbers of sections for each liver. While the reader is convinced that the presence and absence of bile ducts is really related to the genotype of the mouse, the numbers of livers examined is relevant and should be included in this paper (and not referenced to the methods in another paper). In addition, scale bars are standard for histological photographs and these need some.

The introduction (third paragraph) and discussion (second paragraph) both talk about the Jag1/Notch2 knock out, the Notch2 conditional liver specific knockout models made by this group, but do not refer to the Jag1 liver specific conditional knock out by Loomes et al. Discussing this model would improve the logic for seeking to improve the mouse model of AGS given the surprising finding that this mouse model has a mild histological phenotype thus begging the question further about the involvement of Notch2. This would make a much more rounded discussion to include models generated by other groups, improving the lines of evidence not distracting from them.

Minor points:
Introduction 5th paragraph, line 2 "Jag1/Notch2 double heterozygous mice" not "Alagille Syndrome model".
Results: end of first paragraph, "double heterozygote mouse" not "model of AGS"

**********
N.B. These are the comments made by the referee when reviewing an earlier version of this paper. Prior to publication the manuscript has been revised in light of these comments and to address other editorial requirements.