ScoreL 1.1

ScoreL 1.1 is a new original tool for docking, scoring and virtual screening. The fundamental difference of ScoreL from similar software is that ScoreL does not suggest a universal solution for any problem in structure-based drug design. ScoreL is merely a tool you need to adjust to your specific problem to solve it. ScoreL helps you to mine correctly as much information from available experimental data as possible. The more experiments, the more experience.

Successful docking is a necessary condition for good results of scoring and virtual screening. If the most probable position of the ligand is not found correctly, it is impossible to perform correct scoring and, hence, differentiate between active and inactive ligands. The original docking algorithm was developed for ScoreL. This algorithm was successfully validated.
Read more >>

ScoreL uses NScore as a starting scoring function. NScore is a very simple scoring function with as few parameters as possible; yet it allows all the main effects determining the ligand-protein interaction to be taken into account. All NScore parameters have been selected on the basis of the most general considerations, without any adjustment or training using experimental data on ligand-protein interaction. The validation results of ligand docking and scoring with the use of an independent test set of proteins and ligands for NScore were as good as the results obtained using the ICM, GOLD, and GLIDE software, and even considerably better than those with respect to some parameters. The validation results of NScore virtual screening for 8 targets proved to be almost as good as those provided by the GLIDE and Flex software.
Read more >>

Fig.1. The test set for NScore docking is 100 PDB complexes from (Perola E., Walters W.P., Charifson P.S., Proteins 2004;56(2):235-249.); the results for ICM, Glide, and GOLD are from the same paper.

ScoreL offers the user a new method for obtaining docking functions intended exclusively for docking. It was demonstrated that these docking functions can correctly predict the most probable position of the ligand in the active site of the protein in almost 100% of cases. If specially developed docking functions are used for docking, and the subsequent scoring with the use of scoring functions specially developed for scoring is performed for the resultant best position of the ligand in the active site, then the binding affinity of the ligand-protein interaction can be predicted considerably more accurately than if both docking and scoring are performed using scoring functions that have not been developed specially for these purposes.
Read more >>

Fig. 2. The results of validation with the use of a test set of 82 PDB complexes: AutoDock, docking and scoring was performed using the scoring function employed in the AutoDock software; FlexX, according to the results of testing the FlexX software taken from the Web site of this software; ScoreL, a target function specially developed for docking was used for docking, and a target function specially developed for scoring was used for scoring.

ScoreL offers the user a new approach to the development of target-specific scoring functions for virtual screening. According to this approach, an initial scoring function NScore is modified in such a way that the scores calculated using the modified scoring function at the known positions of active ligands in the binding site of the target protein are better than the scores calculated by the same function at any positions for inactive ligands at the binding site. Validation has demonstrated that the selectivity of virtual screening can be significantly increased with the use of the scoring function obtained by this new method.
Read more >>

Fig. 3. The percentage of known actives found (the y axis) vs. the percentage of the ranked database screened (the x axis) for thymidine kinase. Random, random selection; NScore, virtual screening using the NScore scoring function; ScoreL, virtual screening using an active ligand based target function specially developed for scoring.

ScoreL offers the user a new approach to the development of target-specific scoring functions for virtual screening. According to it, information about inactive ligands at the virtual screening target is used to improve the quality of ScoreL prediction for this target. It is important that information about active ligands is not used in this method. Validation has demonstrated that the selectivity of virtual screening can be significantly increased by this new method.
Read more >>

Fig. 4. The percentage of known actives found (the y axis) vs. the percentage of the ranked database screened (the x axis) for cyclin-dependent kinase 2. random, random selection; NScore, virtual screening using the NScore scoring function; ScoreL, virtual screening using an inactive ligand based target function specially developed for scoring.

Other Possibilities

• ScoreL allows the user to adjust a scoring function to the prediction of the binding affinities for the set of active ligands for the target protein, which is very important during the lead optimization process.
• ScoreL allows the user to include constraints similar to pharmacophore in the docking and scoring calculation.
• ScoreL allows the user to predict ligand superpositions. If you have a protein and a small ligand binding to it, but not the structure of the protein, you can take the reference structure as a negative fingerprint of the active site and determine the similarity between the test and reference structures by aligning them

Technical Possibilities

• ScoreL is available for Linux x86 and Windows platforms. The minimum requirement is 1 GByte RAM and 50 MByte of disk space.
• ScoreL can be run simultaneously on two or more computers. This distributed computing significantly accelerates the calculations.
• ScoreL can store the task and result data in the MySQL database, which is very convenient for post-processing the results.
• ScoreL has a friendly graphical user interface, which helps the user in task preparation and post-processing the results.

A Free Demo Version

A free demo version of ScoreL 1.1 is available at www.didiall.com/downloads/.