# Multi-step protocol for HTVS¶

For high-throughput virtual screening (HTVS) applications, where computing performance is important, the recommended RxDock protocol is to limit the search space (i.e. rigid receptor), apply the grid-based scoring function and/or to use a multi-step protocol to stop sampling of poor scorers as soon as possible.

Using a multi-step protocol for the DUD system COMT, the computational time can be reduced by 7.5-fold without affecting performance by:

1. Running 5 docking runs for all ligands;

2. ligands achieving a score of -22 or lower run 10 further runs;

3. for those ligands achieving a score of -25 or lower, continue up to 50 runs.

The optimal protocol is specific for each particular system and parameter-set, but can be identified with a purpose-built script (see the Reference guide, section rbhtfinder).

Here you will find a tutorial to show you how to create and run a multi-step protocol for a HTVS campaign.

## Step 1: Create the multi-step protocol¶

These are the instructions for running rbhtfinder:

1st) exhaustive docking of a small representative part of the
whole library.
2nd) Store the result of sdreport -t over that exhaustive dock.
in file  that will be the input of this
script.
3rd) rbhtfinder <sdreport_file> <output_file> <thr1max> <thr1min> <ns1> <ns2>
<ns1> and <ns2> are the number of steps in stage 1 and in
stage 2. If not present, the default values are 5 and 15
<thrmax> and <thrmin> setup the range of thresholds that will
be simulated in stage 1. The threshold of stage 2 depends
on the value of the threshold of stage 1.
An input of -22 -24 will try protocols:
5   -22     15      -27
5   -22     15      -28
5   -22     15      -29
5   -23     15      -28
5   -23     15      -29
5   -23     15      -30
5   -24     15      -29
5   -24     15      -30
5   -24     15      -31
Output of the program is a 7 column values. First column
represents the time. This is a percentage of the time it
would take to do the docking in exhaustive mode, i.e.
docking each ligand 100 times. Anything
above 12 is too long.
Second column is the first percentage. Percentage of
ligands that pass the first stage.
Third column is the second percentage. Percentage of
ligands that pass the second stage.
The four last columns represent the protocol.
All the protocols tried are written at the end.
The ones for which time is less than 12%, perc1 is
less than 30% and perc2 is less than 5% but bigger than 1%
will have a series of *** after, to indicate they are good choices
WARNING! This is a simulation based in a small set.
The numbers are an indication, not factual values.


### Step 1, substep 1: Exhaustive docking¶

Hence, as stated, the first step is to run an exhaustive docking of a representative part of the whole desired library to dock.

For RxDock, exhaustive docking means doing 100 runs for each ligand, whereas standard docking means 50 runs for each ligand:



### Step 1, substep 3: rbhtfinder script¶

The last step is to run the rbhtfinder script (download sdreport_results.txt for testing):