06.3 Examining Properties and Useful Filters

6.3 Examining Properties and Useful Filters

The first thing to start doing in DataWarrior is adding important properties to table 1, DataWarrior can calculate various properties for your dataset which you can then use to filter your list of compounds down to the ones suitable for docking. To do this click chemistry, then from chemical structure, then click calculate properties.

This will open up a dialog box where you can select the properties you wish to add, the most important are molecular weight, LogP and total polar surface area (TPSA), however the number of hydrogen bond donors/acceptors can also be a useful property to display. Click on the properties you wish to calculate and then click ok.

You will now see that DataWarrior has added all of these properties as new columns in table 1. You can also see that new sliders have been added for each property in the filter area, by scrolling down in this filter area you can now examine the dataset and decide on the ranges you would like to set for each parameter. As you alter the sliders and select different filters you will see the “visible:…” number in the status area change to show how many compounds out of your full dataset you are now viewing (here the full dataset contains 100 compounds and we have no filters active currently hence all 100 are visible).

If we were to change the position of the sliders as you can see in the screenshot bellow where we have altered the LogP value range, the status area now shows “visible: 83” whilst the “total: 100” remains the same. This is because when you filter out compounds in DataWarrior you do not remove these compounds from the dataset. If you want to save the dataset (see section 6.4) with only the desired compounds, then you will need to remove the other undesirable compounds from your dataset (see later).

When filtering your dataset (e.g. by using the sliders) you can invert the filter selection to show all the compounds you are currently filtering out. For the screenshot below we inverted the LogP slider filter to show all the compounds whose LogP values were outside of the filtered range. Note: you will notice the change to the “visible: 17” indicator in the status area.

When choosing filter ranges, for molecular weight you probably want to keep all compounds below 500, and for TPSA anything over 110 can cause problems with oral absorption (drug bioavailability if swallowed). A good LogP range could be between 0-4, however you do not necessarily need to discount compounds just because they have slightly below 0 LogP values. Always discuss with your supervisor or an available expert (such as Dr Swain) if you are unsure on which values to discount, especially for your own designed compounds as these properties are all highly attunable with changes to structure*.* As a general rule of thumb, any extreme values of these key properties (both high and low) are likely to cause issues for drug development. You should research these properties and how they relate to drug development as part of your computational research project.

Another important property to display is the “Nasty Functions” option, this can be found on the “LE, Tox, Shape” tab of the calculate properties dialog box. Nasty functions are reactive functional groups or moieties with known toxicities contradicting their use in drug development. In this dataset there are no compounds containing nasty functions, but it is important to check this and remove these compounds from your list prior to docking.

There are other reasons you may wish to remove a compound from your dataset besides the property filters, such as their conformation being particularly strained (perhaps a 3 or 4 membered ring motif). To delete a compound from your dataset, click on the row belonging to that compound (here clicking on the number 34 at the side will select the whole 34^th^ row of table 1). Next click data, then delete rows, then selected rows. This will delete this entry from your dataset and the status area will change accordingly, showing “total: 99” at the bottom of the page here in this example.

You can remove columns in a similar fashion, if you have calculated a property which you no longer wish to display, or if the original dataset contained information you do not need in your sdf file for docking (such as zinc ID in this example) then you can delete the column by clicking data, delete columns, and then selecting the column you want to delete from the menu in the dialog box that opens. Alternatively, you can right click on the column heading in table 1 and select delete column.

If you have filters active on your dataset and you wish to delete all compounds which have been filtered out then you can easily do this by clicking data, delete rows, then click invisible rows…

This will delete all of the compounds that were being filtered out (i.e. all the invisible compounds). For the above screenshot we altered the parameters of the LogP values, by reducing the range to only show compounds which had a LogP value b