Fig. 5: Summary of score function comparisons. | Nature Communications

Fig. 5: Summary of score function comparisons.

From: Ensuring scientific reproducibility in bio-macromolecular modeling via extensive, automated benchmarks

Fig. 5

Comparison of different score functions (one per column) for different applications and protocols, using the PNear metric as an indication of “funnel quality”. PNear falls between 0 (no funnel or incorrect global minimum) and 1 (the perfect funnel). The lambda parameter indicates the spread on the x-axis and is set to 4.0 in our comparison. Cells are colored according to the color bar on the right, teal is better. Unavailable data is indicated in gray. A The panel uses a “winner-takes-all” comparison: for each protein, the score function with the highest (i.e., best) PNear value (see methods) gets a point. Points are then summed by column, identifying the score function with the most and highest PNear values across proteins, the higher the better. B The averages of the PNear values for each score function were used, i.e., computed over each column. Higher values are better.

Back to article page