E-sol Viewer Panel

E-sol Viewer Panel Features

• Use structures from option menu
• Open Project Table button
• File name text box and Browse button
• Desired endpoint option menu
• Endpoint property name option menu
• Endpoint cutoff text box
• Plot area
• Statistics table
• Log experimental axis option
• Generate PDF Report button
• Save As button
• Close button

Use structures from option menu

Choose the structure source for the current task.

• Project Table (n selected entries)—Use the entries that are currently selected in the Project Table or Entry List. The number of entries selected is shown on the menu item. An icon is displayed to the right which you can click to open the Project Table and select entries. When this option is selected, a Load button is displayed to the right.
• File—Use the specified file. When this option is selected, the File name text box and Browse button are displayed.

Open Project Table button

Open the Project Table panel, so you can select the entries for the structure source.

File name text box and Browse button

Enter the file name in this text box, or click Browse and navigate to the file. The name of the file you selected is displayed in the text box.

Desired endpoint option menu

Choose the experimental endpoint to compare against the predicted E-sol values.

Endpoint property name option menu

Choose the structure property that holds the experimental endpoint values. The list contains the real properties of the loaded structures.

Endpoint cutoff text box

Specify the experimental cutoff used to classify structures for the classification statistics.

Plot area

Displays a scatter plot of the experimental endpoint values (vertical axis) against the predicted E-sol values (horizontal axis), with a best-fit line. The experimental endpoint cutoff and the E-sol cutoff overlay a classification matrix on the plot, dividing it into four colored quadrants with the number of structures in each quadrant labeled. The E-sol cutoff is determined automatically to best separate the classes.

Hover over a data point to display a tooltip with the 2D structure and its endpoint and E-sol values. To select structures, click a data point or draw a closed curve around a group of data points; the corresponding entries are selected in the Project Table. Conversely, selecting entries in the Project Table highlights the corresponding data points in the plot.

Statistics table

Displays statistics for the plotted data, in three sections (Correlation, Classification, and Cutoffs). Hover over a metric to see a tooltip with its definition.

Correlation

• R²: The coefficient of determination for the fit of the predicted E-sol values to the experimental endpoint values; values closer to 1 indicate a better fit.

Classification

Statistics derived from the classification matrix. Each structure is classified as a true positive (TP), true negative (TN), false positive (FP), or false negative (FN) according to the experimental endpoint cutoff and the E-sol cutoff, and these counts are labeled on the plot.

• Accuracy: The fraction of structures classified correctly [(TP + TN) / (TP + TN + FP + FN)].
• Cohen's κ: The agreement between the predicted and experimental classifications, corrected for chance. A value of 1 indicates perfect agreement, 0 indicates chance, and -1 indicates perfect disagreement.
• AUC: The area under the ROC curve, a measure of how well the classification separates the two classes; values closer to 1 are better. Values below 0.8 are colored green, orange, or red as the classification quality decreases.
• p-value: The p-value from Fisher's exact test for the classification; smaller values indicate a more statistically significant separation. Values greater than 0.05 are colored red.
• Specificity: The proportion of actual negatives that are correctly classified, or true negative rate [TN / (TN + FP)].
• Sensitivity: The proportion of actual positives that are correctly classified, or true positive rate [TP / (TP + FN)].
• Recall: The same as sensitivity [TP / (TP + FN)].
• Precision: The proportion of predicted positives that are correct, or positive predictive value [TP / (TP + FP)].
• F1 score: The harmonic mean of precision and recall [2 × (Precision × Recall) / (Precision + Recall)].
• TP/FP ratio: The ratio of true positives to false positives [TP / FP].

Cutoffs

The cutoff values used to classify the structures: the experimental endpoint cutoff (from the Endpoint cutoff text box) and the E-sol cutoff (determined automatically).

Log experimental axis option

Select this option to plot the experimental axis on a logarithmic scale.

Generate PDF Report button

Click to generate a PDF report containing the plot and the statistics tables. A file selector opens, in which you can navigate to a location and name the file.

Save As button

Click to save the correlation plot as a PNG image. A file selector opens, in which you can navigate to a location and name the file.

Close button

Click to close the panel.