In recent years, reproducibility of ‘scientific’ publications has become quite a topic. See a recent essay Five selfish reasons to work reproducibly by Markowetz in Genome Biology (2015, 16:274). There are numerous reasons why reproducibility could become an issue at all in science. What I have continuously strived for in my scientific career, however, is to ensure that my published results are reproducible. As a concrete example, I created a dedicated section titled DSSR-NAR paper on the 3DNA Forum that provides full details (scripts and data files) so that any interested parties can rigorously reproduce the results reported in the DSSR Nucleic Acids Research (NAR) paper.
In my support of 3DNA for over a decade, the #1 issue I experienced is undoubtedly vague (non-reproducible) questions. For example, I have recently been asked via email why the 3DNA find_pair/analyze programs miss “some basepair … even though it is in the pdb file”. Without access to the PDB file to reproduce the problem, however, I cannot provide a concrete answer. In an effect to prevent ambiguous questions, I made the following explicit point in the “Registration Agreement” of the 3DNA Forum (no. 2 on the list):
Be specific with your questions; provide a minimal, reproducible example if possible; use attachments where appropriate.
The #2 issue is receiving 3DNA-related questions privately instead of on the intended public 3DNA Forum. I turned off “personal messaging” to receive private messages on the Forum long time ago, yet I have kept receiving questions via emails. In several locations on the 3DNA Forum, I have made this ‘public-question’ policy crystal clear:
Ask your questions in the public 3DNA forum instead of sending xiangjun emails or personal messages. (no. 1 on the ‘Registration Agreement’)
Please be aware that for the benefit of the 3DNA-user community at large, I do not provide private email/personal message support; the forum has been created specifically for open discussions of all 3DNA-related issues. In other words, any 3DNA-associated questions are welcome and should be directed here. Presumably I’ve made the message simple and clear enough to get across without further explanation. (in ‘Site announcements » Download instructions’ and ‘Downloads » 3DNA download’)
In response to the many 3DNA-related questions that still keep coming via email, I created the following entry of Canned Responses in gmail:
Thanks for your interest in using 3DNA. Please be aware that for the benefit of the 3DNA-user community at large, I do not provide private email support; the 3DNA Forum (http://forum.x3dna.org/) has been created specifically for open discussions of all 3DNA-related issues. In other words, any 3DNA-associated questions are welcome and should be directed there. I monitor the forum regularly and respond to posts promptly.
I look forward to seeing you on the 3DNA Forum (http://forum.x3dna.org/).
Overall, I’ve learned from experience that addressing reproducible questions publicly does the best for the 3DNA community. Users can register with personal (free) email address, and post simulated data to illustrate the problem at hand. Moreover, questions on the Forum have always received quick responses. Over time, the Forum has served as an archive that everyone can benefit from.
