Tutorial - How to use the TransClust web interface


Cluster protein sequences by using BLAST/FASTA files

  1. Navigate to the TransClust web interface site
  2. In the select Type section, select "BLAST & FASTA file". Further options will be displayed.
  3. Download the BLAST file as well as the Example FASTA file
  4. Click on the file input field behind "FASTA file" and "BLAST file". A file chooser will open. Select the example files downloaded in the previous steps.
  5. Leave the options under "Advanced..." the way they are and click on the Upload files button.
  6. On the next page, you will see a graphical representation of the similarity distribution. In this histogram the x-axis represents buckets with ranges of similarity. The number of buckets can be adjusted with the slider above the chart. The y-axis shows the number of object-pairs with the similarity contained in the corresponding bucket. Clicking on the "Similarity file preview" tab will show the first 100 rows from the created similarity file.

    Below the chart/preview you can find further clustering options. On the "Use multiple Thresholds for Clustering" side provide the lower and upper bound for the range of thresholds to be clustered. Set the lower bound to 10 by moving the left part of the slider or by entering "10" in the text field below "From:". Set the upper bound to 60 accordingly.
  7. Set the number of steps to 5 by entering it in the text field below "# Steps:". The Stepsize will be computed automatically.
  8. Click on "Advanced..." to enter further options. To evalutate the clustering you might want to provide a gold standard file. Download the this gold standard file. Click on the file input field below "Goldstandard file:". Leave the other options the way they are and click the "Cluster" button below the Merge objects part.
  9. Wait for the clustering process to finish. You can also receive an email, when the clustering is done by entering your email in the corresponding text-field in the top of the page and pressing the submit button. Or you bookmark the page by clicking on the "bookmark" link and return to check, if your results are ready.
  10. After the process is finished, the results are displayed. Use the tabs to switch through the different outputs.