Help topics for 2DB

You may find 2DB at several locations. The most reliable is however 2DB.de.ms. You may also try biolnk.com. Regardless of which of the server you choose for download, you need to go to the download section in order to choose from the available versions.

For using 2DB locally on one computer, you need a webserver, a database and PHP. All of this comes convenienly packaged in XAMP(LINUX), MAMP(Mac) and WAMP(MS Windows). Merely installing the application is not enough, but comes close.

You may find 2DB at several locations. The most reliable is however 2DB.de.ms. You may also try biolnk.com. Regardless of which of the server you choose for download, you need to go to the download section in order to choose from the available versions.

For using 2DB locally on one computer, you need a webserver, a database and PHP. All of this comes convenienly packaged in XAMP(LINUX) and WAMP(MS Windows). Merely installing the application is not enough, but comes close.

This tool provides more thorough calculations than would be possible with the web interface of 2DB. Currently, no statistical analysis is performed and the results are intended to give you only a first insight into potentially differential protein expression. Distribution patterns of proteins/ peptides over the fractions of an experiment are meant to help you find proteolytic processing of proteins. Several modes of analysis are possible with this extention to 2DB:

• Displaying distribution patterns of proteins in your experiment
• Displaying the distribution patterns of the (associated) peptides
• Label free quantitation
• Based on equally named fractions
• Based on fractions (spots/bands/..) pooled for several experiments
• Based on identified proteins/ peptides
• Future work
• Inclusion of PTMs
• Quantitation with one or multiple labels
• Statistical analysis

Developed by Jens Allmer.

This interface provides the possibility for label free quantitation. Be aware, that no statistical analysis is available at this point. In order to have your results somewhat more sound, it pays to set peptides as loading references which you are sure will not vary in their expression level under the varying conditions applied in the experiments under investigation. If several peptides are selected they will be averaged. Peptides that are present in a fraction are preferentially used. This loading control is used to adjust the protein and peptide abundance thus leading to better resutls.

---

Setting peptides (standard/ loading reference)
The second list from top on the left contains peptides which are shared among all experiments that were selected. Select as many peptides, which shouldn't vary in their expression level, as possible (hold Ctrl and click) then click Add to set these peptides as loading references. These peptides are transferred to the last list on the left. Selecting peptides in that list and pressing del allows to remove peptides from that list.
Display all proteins
By default proteins and peptides present in less than all of the experiments are hidden. Checking this box will display these proteins and peptides as well.

---

Pool
In order to use this mode of quantitation, it is necessary to set the appropriate pools using the 2DB interface, first. Only fraction(s) that are pooled with fraction(s) from other experiments are then quantitated. Understand that this involves a significant amount of work so it may be better to plan naming of your fractions.
Name
If the fractions of the experiments that shall be compared are chosen such that the corresponding fractions in different experiments have the same name, they can be quantitated directly and pooling need not be set. Pooled fractions (per experiment) are evaluated however. This means that if two or more fractions in one experiment are pooled, their identified peptides are also pooled for this amalysis.
Protein/Peptide
If neither the names correspond nor pooling has been set, a global comparision for the experiments will be made. All fractions will be pooled in regards to the protein and peptide content. Then the result will be quantitated.
Auto
If pooling has been set that method is used. Following this name and lastly Protein/ Peptide methods are tried automatically.

---

Proteins View
Selecting a separation in the first list on the left will make it the reference and it will be the first column in the table containing all results. The protein results can be accessed by selecting the protein tab on the right. By copying and pasting the results to a spread sheet application, further analysis can be performed.
Peptides View
Works essential like Proteins View.

This interface provides the possibility for viewing the distribution of identified proteins over the fractions of an experiment. Proteins can be selected in the tree view on the left. All fractions containing the protein will be indicated in the upper graph on the right. The y-value of the protein shows the percentage of the overall TIC (all TICs of all supporting peptides summed) in the respective fraction. The lower graph shows the same information for the associated peptides. Initially the first peptide following the selected protein in the tree view on the left is displayed. This selection can be changed by clicking on a differnet peptide in the tree view.

---

If a protein has only one peak, you have a well defined fraction.
If a protein has several peaks (close to each other), your protein eluted to several fractions. The protein content can be estimated by the peak height. If a protein has several peaks (distant to each other):
You may have a protein which can be identified by peptides that are shared among other proteins. This will lead to false positive identifications in other fractions (look for similar proteins). You may have a protein which has been proteolytically processed and you pick up the processed products. Verify that the associated peptides are not part of other proteins (use search facility of 2DB and type the sequences).

---

A separation is something like a Gel or an LC run. In order to apply pooling the appropriate separations need to be selected first. This means that the fractions that ought to be pooled need to exist in the two separations.
It is possible to pool several fractions on one separation by selecting the separation twice.

• Select a separation in the left list (Separation 1)
• Select a separation in the right list (Separation 2)
• Click submit and select the appropriate fractions

Pooling is a very simple process but only available for two separations at a time. It is also only possible to pool two fractions at a time. In order to pool more fractions, pairs have to be paired repeatedly. The same can be done for multiple separations. A pool can be deleted by clicking the del button under the Pool list.

• Select a fraction from the left list
• Select a fraction from the right list
• Click add button under the Pool list
• Repeat until all pools have been created
• Finalize by clicking Submit

All protein identifications can be significant in regards to your settings. Many settings, apart from the Thresholds that are specified in the software section influence the detection of protein from the data in the database.

Options that influence the detection of proteins.

• Use of pooling
• Rechecking Scores
• Check Organism
• Thresholds

Too complex for now.

Please refer to the Section on pooling for more information.

In the software section of the database thresholds can be set for the import of results from MS/MS identification software and for 2DB.
As long as the criteria you are currently using do not differ from the ones used for import there will be no effect by using this option.
Clearly, the only time that this feature will be of use is when more stringent thresholds have been set after import of the data.
Only for the significance criteria for 2DB, is it possible to play into both directions (depending on your current settings of course).

This feature is currently not enabled.
At the moment only proteins that come from sequences connected to the organism, specified for the experiment containing the data, are used for establishing protein identifications.
In the future this feature will enable the use of all sequences stored in the database regardless of which organism they are connected to.