Referee 1's review:

**********
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.
**********

Comments to Authors:

Review of manuscript "A high though-put reverse genetic screen identified two genes involved in remote memory" by Matynia et al.

This manuscript describes a study designed to screen and validate various mutant mice for phenotypic differences in the conditioned fear test that are test-delay dependent. The paper is very well written and describes a very important and powerful approach to identify genes for conditioned fear phenotypes. There are only a few suggestions that should be addressed to improve the overall quality of the paper.

The paper is well designed and clearly well executed. The use of the Z-score is appropriate for the first phase primary screen, and is validated in the secondary screen. The work with the different inbred strains first appears to be a valid approach to try to ensure similar levels of 'fear' following training. The follow-up secondary experiments are critical and the results are very compelling. There is only one potential issue that should be addressed in the paper. The authors tested the two mutant lines that showed the interesting remote phenotype once backcrossed onto a B6 genetic background and the phenotype was not present, demonstrating the power of modifier genes. However, it is not clear if the during this experiment if the authors used the multi-shock or single shock training protocol. If I followed the methods correctly, in the primary and secondary screen these two mutant lines were trained with single CS-US pairings because of their genetic background. Yet, other mutants on the B6 background were given multi- CS-US pairing in the primary screen. Therefore, were the two 'remote-memory' mutants given single or multi CS-US pairing when backcrossed onto B6? If they were given multi CS-US pairings, then it would be essential to know what the memory performance is when the two 'remote-memory' mutants are given multi- CS-US pairing on their original background. It is possible that if they were given more CS-US pairing on the original background that the remote-memory phenotype might also been absent.

Other minor points.
First, the suppression ratio should not necessarily be given too much weight in the overall analysis; since there is no control data to know how much suppression of behavior there is when a mouse is trained with NO shock then tested later to determine the suppression of activity simply from being placed in the apparatus on two occasions.

It would have been interesting to see if these mice also showed normal CS fear over these test delays; Any thoughts or predictions from the authors on this type of fear memory?

Lastly, it does not appear that the figure legend for figure 1 has a description or listing for panel 'c'. This reviewer may have just missed it, but I think it is missing.