This tutorial will demonstrate how to analyze a dual-channel high-throughput screening experiment using web cellHTS2.
Step 1: Download test data
Demo data sets
As a first step, unpack the downloaded files into a directory for upload to the web cellHTS2 interface [link].
Step 2: Start the web application and select type of experiment
After starting the web application, select "Dual"-channel data analysis and name the channels (for example: "channel 1" and "channel 2"), then confirm the selection by clicking 'next':
Step 3: Upload raw data files
Upload the data files one-by-one or pack all datafiles ("RA01D1.TXT" to "RB03D2.TXT") (from the test demo folder) into a zip file and upload the generated zip file. Upload the files with the "traditional" data file uploader (click the button "browse" to do so). Do not use the Advanced File Uploader. There is a seperate tutorial on how to use the advanced file importer here .
After uploading to the webtool, a table with all the valid datafiles and extracted parameters will be shown.
Most uploaded files were correctly assigned, except the file "RC03D2.TXT" which will be automatically interpreted as channel 3 (because RA=channel1 and RB=channel 2). Edit the channel number by clicking on the channel field of this file. An editline will appear and can be changed to the correct channel number. To finish, press return or click outside of this field.
Change the channel name to "2" and click on the next button.
Instead of editing the file name parameters, there are two more options. Let's try some additional options. Use the "back" button to go back to the datafile upload step (step 2).
Option 1: Use a regular expression to extract filename parameters (expert options).
In order to see the full power of it, lets first analyse the filenames if there is a pattern:
R<CHANNEL NUMBER ENCODED AS A OR B OR C LETTER><PLATE NUMBER>D<REPLICATE NUMBER>_.TXT
Imagine the situation that we would need a different layout:
R<PLATE NUMBER ENCODED AS A OR B OR C LETTER><REPLICATE NUMBER>D<CHANNEL NUMBER>_.TXT
This would be written in web cellHTS2 as (for further explanation review the manual)
Paste this string into the apply regexp textfield and press the update parameters button.
Option 2: Upload an existing plate list file.
This option might be handy if, for example, a plate list file from similar experiments exist and has been saved.
1. Delete the pasted regular expression and press "update parameters" to not use any regular expressions.
2. Upload the plate list file included in the demo data set.
Do not use the plate list parameters from the plate list file just uploadedbecause it is only for demonstration purposes. cellHTS2 will produce errors if this file is used for analysis. Follow the instructions just below before moving to the next step:
3. Delete entries by pressing "Drop all entries" afterwards (because the platelist file entries are only to demonstrate how to use and not to actually for analysing the testdata). Repeat the first part of the Step 3 until the "Change the channel name to "2" and click on the "next" button.
Step 4: Configure plates
This step demonstrate how a plate configuration file will be uploaded, or being created manually. Plate configuration files are necessary to indicate which wells should be considered as positive, negative or other type of controls.
1. Please locate "Plateconfig.txt" provided in the demo data set.
2. Upload the file, the following plate configuration layout will be shown
You can now change the displayed option and select a well type out of the drop down menue (positive, negative) and in the upper drop-down menue and click on the plate to mark wells with those types. The standard well type is the "sample" one (grey) which will be used if you click on a well defined not as sample again.
If the "all" plate has been chosen, changes will be applied to all the plates in the set. For any other plate from the lower drop down menue all positive, negative or other well-changes will be applied only for all the plates with the same plate ID.
For instance if some wells get marked positive on the plate and replicate combination plate 2, replicate 2 (PL_2_REPL2) they will automatically be marked on plate PL2_REP1 because of the same replicate number whereas if a contaminated well on any other plate than the "all" plate gets marked, it will only be set for this one specific plate. Contaminated wells only "overlay" other wells so if they get clicked again, the underlying welltype will appear.
After testing all the features, delete the complete layout by clicking on "x" on the "all" plate and mark the following wells on the "all" plate for a sound analysis
If at least one well has been marked with any available type click on the "next" button now.
Step 5: Choosing options for statistical methods
For a full documentation on different standardization and normalisation methods see here or click on the Help links provided with every normalization method.
Use the following settings for the demo data set (or try other options):
Step 4 = median (standard)
Step 5 = NO (standard)
Step 6 = multiplicative (change)
Step 7 = by experiment (change)
Step 8 = mean (standard)
Step 9 = yes (change)
apply viability function = paste the following function:
function(r1, r2, thresh = quantile(r1, probs = 0.05, na.rm = TRUE)) ifelse(r1 > thresh, log2(r2/r1), + as.numeric(NA))
Step 6: Uploading additional data
On this page the Gene Annotation file called "GeneIDs.txt" (from the demo archive provided with this tutorial) should be uploaded through the form labeled "step 10". If everything went fine here a blue text will show up saying "Annot file successfully uploaded".
Next step will be uploading the provided Description file (Description.txt) to the upload field in step 11. If uploading of the description file was successfully the site should look like this:
On this page in the description form the experimenter and celltype fields are still empty after uploading the description file which can be filled in now (Description file is completely optional so you do not have to add any information). Please note: If any fields in the description editor get updated or filled in please click on the "Create/Update" button or your changes will not be permanent in cellHTS2.
Last but not least scroll down a bit this page and click the button "Start analysis run"
After clicking it switches to a new page where a progressbar and a textbox can be seen which shows the current step the R part of the program is at.
If the progressbar is at 100% a popup window should appear with the results packed as a zip file.
after downloading the job it can be extracted and the contained index.html can be opened in a webbrowser.
Please Note: for some versions of this tool instead of a progressbar and directly streaming of the results via HTML, results will be sent to an email address which can be provided at this step.
Step 7: Saving and restoring analysis sessions
To save a session for later analysis, click on the "back" button on this page so that we are again at the Step 5 page. Click on the link: "Save this session"
a popup window will appear to download zip file including all raw data and options.
The saved analysis can be restored on the front page of cellHTS2 (Step 1) by selecting the "Load previous session:" upload form. After restoring a previously saved session, the user can proceed to Step 6 of this tutorial where the analysis can be re-started or changes to parameters can be introduced. To do this now click on "Start new Analysis" now to go to step 1 where the downloaded file can be uploaded.