Wednesday, June 13, 2012

Reviews of Martin's 2nd PLoS ONE article

The reviews for Martin's second PLoS ONE paper, which was submitted May 4, came back Monday.

My gut reaction is "Nothing we can't handle".  In particular the quantitive aspect specifically mentioned by the editor as we already have a paper on this.

The most critical change I think is adding 2-3 paragraphs in the introduction that summarizes the model, the assumptions behind it, and its use to interpret the results by us and others (i.e. very briefly summarize the papers who have used the previous BioFET-sim papers or the original paper by Mads Brandbyge and co-workers).  Yes, there are approximations in the method but given the accuracy with which key experimental parameters are known at present for these sensors, more accurate methods are not really warranted.

We also need to be more explicit about the availability of the code as requested by the second reviewer.  Perhaps we should simply make it available on Github and stop messing with the distribution ourselves?

BioFET-SIM Web Interface: Implementation and Two Applications

Dear Prof Hediger,

Thank you for submitting your manuscript to PLoS ONE. After careful consideration, we feel that it has merit, but is not suitable for publication as it currently stands. Therefore, my decision is "Major Revision."

We invite you to submit a revised version of the manuscript that addresses in particular the important points raised by ref 1 on the quantitative aspects of the results of the web application.

We encourage you to submit your revision within sixty days of the date of this decision.

When your files are ready, please submit your revision by logging on to and following the Submissions Needing Revision link. Do not submit a revised manuscript as a new submission.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

Please also include a rebuttal letter that responds to each point brought up by the academic editor and reviewer(s). This letter should be uploaded as a Response to Reviewers file.

In addition, please provide a marked-up copy of the changes made from the previous article file as a Manuscript with Tracked Changes file. This can be done using 'track changes' in programs such as MS Word and/or highlighting any changes in the new document.

If you choose not to submit a revision, please notify us.

Yours sincerely,

xxx (name removed upon request)
Academic Editor

Reviewers' comments:

Reviewer #1: Review of "BioFET-SIM Web Interface: Implementation and Two Applications"

As a true admirer of the web-based computing and simulation through, I can not overstate the importance of web-accessible, easily usable, computational tools for designing and interpreting experiments. Over the years, I have contributed a series of tools and therefore I understand the effort it takes create, document, and distribute such an application. Therefore, I read the paper with significant interest.

I have several questions regarding the physical model that needs to be clarified:

(1) The authors calculate the conformation and charge of proteins by downloading data from PDB and using classical formula for calculating pH-specific charges of the residues. I wonder if the calculations are correct when a metal surface is present that creates an image charge - changes the conformation significantly. I wonder how this effect is accounted for.
(2) Salt screening appears to be included through Sorensen's approach through Eq. 7 (Is this true)? The authors do not describe clearly how the screening model was adapted for this particular application? What are the limitations of the approach?
(3) What about pH-dependent surface charge associated with SiO2 and Al2O3 through site binding theory?
(4) The model for transistor part is exceeding simple. I see that there is no distinction between inversion, depletion, and accumulation modes of operation. Even simple SPICE model could be very useful.
(5) Where is the reference electrode and how is the effect is accounted for in the equations? People often use the reference electrode to bias the transistors and I see no mention of reference electrode in the paper.

Frankly, my feeling is mixed: If I recommend publication, people will have a toy model to give qualitative insights into the biosensing problem. The string of software linked together is also useful - people find it difficult to use them separately. However, the reason I feel uncomfortable recommending the publication is that despite the details giving the appearance of complicated analysis, the results are likely to be quantitatively incorrect. I worry about it particularly because the authors in their work did not project a nuanced understanding of the problem. The authors should consider rereading works by Berg, Manning, Landheer, Nair/Alam, etc. and rewrite the paper to highlight the assumptions, limitations, and the context. Also, a review of existing biosensor simulators in reference to the current model will be useful. I will recommend publication if these revisions are made, references are updated, and assumptions/limitations are clearly hightlighted, .

Reviewer #2: First of all, I have to say that my working field is not at all close to Biosensors, so I'm not doing any comments about the theory behing this web application hopping that other referees will. However, I do work with protein structures and web applications that handle them, so I still feel I can contribute to review this paper.

The manuscript describes a web application devoted to simulate the change of conductance in a nanowire when a biomolecule is bound to it, depending on the orientation of this biomolecule. According to the authors, this process used to require some complex steps that they have reduced to a couple of clicks and a few parameter adjustments, all in a very visual way. The authors also provide two examples. In the first one, they demonstrate how different orientations of the biomolecule on the nanowire influence the final result due to the different positioning of charges. In the second example, they propose a new explanation for experimental results from a previous work.

I think this web would be very helpful for those working on this topic.

General comments:
This work seems to belong to a very specific field and to be directed to people with extended experience. It would hardly be used by someone out of that field. Taking that into account and Plos ONE being a very general journal, I think that the introduction is weak and does not properly present for neophytes the context, what the application does, the meaning of the outputs and how to interpret them.

Elements that apparently have important influence in the real behavior of the system, like the biolinker and the oxide layer, are almost not mentioned, they are just named in the introduction. How do they influence the system? How are they taken into account in the calculations? This should be described in the article.

Although the definition of an open source software does not mention free download availability, it is commonly agreed that distribution should be without restriction. "Can be obtained from the authors" is already a restriction and authors should reconsider the open-source label.

Specific comments:
In the web application, it would be helpful to place next to the input boxes a recommended maximum and minimum value for each option. Using a "<SELECT" html tag instead of an "<INPUT" tag for the "NW Type" field would make sure that only one of the two possible options will be selected.

Regarding the Jmol applet and the possibility to adjust the orientation of the biomolecule, it would be helpful to be able to save the new position so it could be used for other runs, ensuring reproducibility.

According to the web page, CTL + Left-mouse key is to translate the view. This key combination did not work for me, however CTL + Right did.  This should be corrected.

Page 5 line 56 -  p6 L7, authors assume that "all biomolecules are oriented in the same way", which makes sense if all are bound at the same residue and are influenced by the same charges, but authors do not provide an explanation or a reference to support this.

Page 8 line 15, authors use the word nanodevice like it was a perfect synonym of nanowire, is that correct?

In the same page line 31, Figure 1S is mentioned. Is it Figure 14 of the supporting information?

In page 9 line 47, I recommend to remove the word "completely" from the sentence "completely platform independent". The Jmol applet did not work properly in my Fedora 12 workstation (and Jmol does)

According to section S1.2 "all positions in the antibody sequence where mutated to Glycine". How the results of such an unrealistic molecule can be applied to a real experiment where amino acids of the antibody do have charges?

Figure 12 presents 2 graphics, to improve they comparison both should have the same maximum value for the ordinate axis.

[NOTE: If reviewer comments were submitted as an attachment file, they will be accessible only via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]


baoilleach said...

When I see "Can be obtained from the authors" I understand "We did not spend the time to release this properly". :-)

But also it prevents the code from entering the scientific record. I've just read a paper from 2004 on detecting the symmetry point group, and when it said "Code available from the author on request", I thought to myself, "Really? 8 years later?" Why didn't the author just attach the code to the paper as supplementary information?

Jan H. Jensen said...

I agree with you. The question is whether we set up an automatic service ourselves (like we do with PROPKA) or use something like Github. I am leaning towards the latter ...