Welcome to SpectConnect!

Analyze your GC/MS data in three steps...

Step 1: Extract putative components using AMDIS

Click here to show more detailed instructions on how to use AMDIS.

Step 2: Use SpectConnect to find conserved components and biomarkers

Fill in the fields below and click the submit button.

Number of conditions:       Maximum number of samples per condition:


Condition 1 name:

  .ELU file #1:
  .ELU file #2:
  .ELU file #3:

.MSL lilbrary file for target identifications (optional):

Email address to receive notification/results (optional):

Step 3: Analyze the results that you receive from SpectConnect.

Click here for some tips on interpreting the results that SpectConnect gives you.

NOTE regarding recent server slowness (7/26/2010).
Lately, some jobs have been taking longer to return due to individual users submitting large numbers of large-sized jobs that monopolize the queue. In order to restrict this, we have implemented a system that restricts users to one job at a time. We hope this will allow for more equal sharing of our limited but freely shared computational resources; those who submit only a handful of jobs concurrently or individual jobs with less than 3000 peaks per ELU file should detect little difference in wait time for results compared to before our recent difficulties. If you experience any further problems or have questions regarding this prioritization of jobs, please contact us at marksty <at> gatech <dot> edu.

NOTE: Here is all of the source code needed to run SpectConnect so that you can use it on your own computer/server via a command-line interface. This will require Python and a C compiler. It has some software library requirements that are described at the home page for Gemoda, a software package used by SpectConnect. In addition, it requires a relatively recent version of Python and SciPy, which may itself have some requirements such as NumPy. Though documentation is sparse, using a "--help" flag will give a starting point for using this command line interface.

If you'd like some sample files, feel free to download this file and this file to your computer (right-click each link and choose "Download" or "Save link target as"), adjust the parameters above, and then use the "Browse..." buttons above to find them and give them back to SpectConnect as input.

This page was written for W3C standards-compliant browsers; browsers that do not follow these standards (e.g., IE before 6.0) may display suboptimally. Thanks for understanding.

Note that your input files and libraries are not stored; once your results have been generated, these files are deleted to ensure your privacy. Results are saved for a set period of time to allow you to access them and are automatically deleted after that time.

Questions? Please don't hesitate to email us at marksty <at> gatech <dot> edu