Tools to Ease the Choice and Design of Protein Crystallisation Experiments

Round 1

Reviewer 1 Report

The manuscript is well written and describes an important set of software tools to enable structural biologists to better generate crystals. 

Reading the manuscript had me highlighting all of the things that I wish I had in terms of software in the Crystallization Center that I run.  I am excited to see these software tools described and look forward to being able to make use of some of them!

I have a few minor questions/comments for the authors:

1) line 18: I think it would be worthwhile to mention methods other than vapor diffusion; even though most people use the vapor diffusion method for crystallization their samples, there are others such as microbatch under oil.  Optimizing from these other methods is sometimes complicated by the need to switch to vapor diffusion. 

2) line 82: I think it would be helpful to define and briefly describe MARCO when it is first mentioned.

3) line 113: What type of clustering is used?  Hierarchical?  It might be helpful to more fully describe the distance metric used for that clustering.  Similarly, what is the 'Score' listed in Figure 1?  Is there a cutoff used for the clustering?

4) line 139: Is there a reference for the distance metric used by C6?  If not, this could be more clearly laid out.  If there is, this section would be a good place to list it.

5) Figure 9: Do the asterix marks indicate better quality crystals?  It looks like there are 3 crystal categories listed.  What are the features that correlate with scoring a 'higher quality crystal'?

6) line 373: Would it be appropriate to reference the doi for the MARCO implementation pipeline that is available?

7) A general question is whether there are tutorials available for the various software options described.  If there are, it would be great to mention those.

8) The tools described for the recipes (shown in Figure 4) are very useful.  

Note all of these items are minor questions - I think this paper reports important contributions to the field and recommend publishing.

Thanks for the opportunity to review this manuscript. 

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

"Tools to ease the choice and design of crystallisation experiments" by Rosa et al. is a well written description of some very sophisticated software tools developed to aid the initialisation and exploration of the protein crystallisation space for any given protein target.

The paper describes the development of a ‘tour de force’ of tools that any modern protein crystallographer would benefit from using and I recommend this be published in Crystals as the aims and scope fit the journal well and should be of considerable interest to protein crystallographers.

Highlights include:

the software allows the comparison of the chemical space each crystal screen explores. While much of the screen data is easily available from each distributor this software conveniently brings it together in a readily accessible format so that with a simple click of a button a market-wide analysis can be carried out in a few seconds so that the user can continue directly with selecting appropriate starting points. the program enables a user to output simple recipes whereby any condition can be made from stock reagents. This takes away the often boring and time-consuming calculations required for even a single condition to be made and tested —thus increasing the potential experiments that a user may try to optimise and should also reduce potential errors. The Phase Space visualisation module is a unique tool that enables positive correlations of chemical classes on crystallisation to be teased out of large experimental datasets of target proteins in various crystal screens. This is important as users typically carry out a large number of varied commercial screens with varying success and due to the large number or variables are sometimes unable to identify correlations assisting in crystal formation. See3 enables the crystallisation hits to be ranked and summarised by imaging and scoring and enables users to move around their (often) large datasets to summarise the successful hits neatly and succinctly, enabling the user to focus on the chemical spaces that work best for their target. The automatic optimisation and manual optimisation functions are potentially exciting. It is often very hard to teach those learning how to make crystals how to optimise their hits as each protein target is different and a number of factors need to be taken into consideration. This tool will be useful to help guide those learning this process.

The paper is well written, largely free of typographic errors and contains a number of suitable and illustrative figures. My only suggestion is to include protein in the title as the work has been developed focused on macromolecular protein crystals suitable for X-ray methods.

Small typographical corrections are:

Line 48 (insert) “…  but there are few software tools available which that help….”

Line 183 (replace) : ‘Sptlabtech’ with “ SPT Labtech”

General comment: there are a large number of ‘en dash /hyphens’ used in the text that should be corrected to ‘em dashes’ in the final manuscript.

 

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

In this manuscript by Nicholas Rosa et al., the authors represent the software developed in the C3 lab to choose suitable screens for crystallization trials, to analyze the results, and to design of conditions for optimization of crystal growth.

The See3 program with incorporated an artificial intelligence tool MARCO is application for the viewing, scoring and optimization of crystallization experiments and the C6 program is a tool for the comparison of crystallization screens. Both C6 and See3 speeds up the process of the rational search for crystallization conditions and analyzing the results of crystallization trials. Websites have a friendly interface.

The manuscript describes User Manual in rather detail. In some cases, a too detailed description of the processes given in my opinion (for example, page 8, lines 173-175).

Author Response

Please see the attachment

Author Response File: Author Response.pdf