Does Anyone Have A Lims That Handles Emails?

Question

[Jeremy Leipzig]

Typically I will start a sequence analysis project in which users later email me crucial details, which I have to record somewhere in the project folder.

This is getting out of control as the number of project folders expands. It seems obvious that I need some kind of LIMS system. It has to be loose since I do so many different types of analyses. Perhaps a wiki?

One feature I would need is a filter which reads my email and collects relevant messages based on keywords or some way to add easily add emails to a project, like a issue tracker would offer (we have JIRA). I really need some way to easily designate emails as being relevant to a project so I am not relying on copy-and-paste.

How do you manage this problem?

[Istvan Albert]

I was entertaining the thought of writing a small LIMS for sequence tracking and now your feature will be on my TODO list. At first it sounds superfluous but then since so much of the discussion happens via email, it would actually be very neat way to be able to easily associate and store that discussion with the project. The way to do it would be to either create a sendmail configuration that reads the incoming mail and uploads it to a URL or regularly parse the inbox from the LIMS.

Answer 1

[Istvan Albert]

There is an easy way to add this if your LIMS supports adding comments.

You can set up a cron job that periodically reads the inbox, parses the emails of interest then uploads them to the right destination.

[Jeremy Leipzig]

yeah I might use this plug-in to twiki

I suppose I could add the right subject tags to classify emails using Outlook scripting or something http://twiki.org/cgi-bin/view/Plugins/MailInContrib

Answer 2

[Sb3]

Setting up a cron job and using awk/perl or whatever would not be my choice; what e-mail client are you using - does it have "rules" or equivalent ? Keyword searching is also unreliable - you can't rely on e-mails always being spelt correctly, using the same terms, etc etc.

I would change the business process and set up a CMS, or a small bespoke website.

[Istvan Albert]

the goal is not organize one's email but having the option of attaching past email conversations to a web enabled project

[Sb3]

Thanks, but I still think my suggestions are valid.

"One feature I would need is a filter which reads my email and collects relevant messages based on keywords".

[Jeremy Leipzig]

This would work in a perfect world. I doubt a senior level PI could be easily persuaded to use such a system. Because I am solely responsible for the analysis, their just response would be "why are you making me respond on your website when you asked me a question about fragment lengths? Here is my reply, record it where ever you please." Basically I can only hope they don't munge an identifier that specifies what experiment we are talking about.

Answer 3

[Jarretinha]

This is an interesting situation which we all found at least once during our bioinformatics life. The simple solution is that of Istvan. Just set up a bunch of shell scripts with you favorite job scheduler and parse the mails & cia with sed/awk/grep. This approach is secure, robust and can be specifically tailored to your very needs.

Another solution is to setup a CMS and configure it as your LIMS. You can find some examples that use Plone as a LIMS. An example is the now obsolete Bika LIMS or Cerium Labs. You can use Plone4Bio as a LIMS too. The great advantage of it is that people themselves will do the parse part. They can add data to the filetree online, safely. And is trivial to add a issue tracker, a wiki, a board, schedule, etc. The problem is: Plone uses a server. So, security is a concern.

Of course, there are other CMS with different flavors, like Apache Lenya. You just need to adapt them to your needs.