Changes to images can create misleading results, especially in research data collected as images. It may, however, be legitimate and even necessary to edit images. We ask authors to declare where manipulations have been made.

• Specific features within an image should not be enhanced, obscured, removed, moved, or introduced.
• Original unprocessed images must be provided by authors should any indication of enhancement be identified.
• Adjustments to brightness or contrast are only acceptable if they apply equally across the entire image and are applied equally to controls, and as long as they do not obscure, eliminate, or misrepresent any information present in the information originally captured.
• Excessive manipulations, such as processing to emphasize one region in the image at the expense of others, are inappropriate, as is emphasizing experimental data relative to the control.
• Nonlinear adjustments or deleting portions of a recording must be disclosed in a figure legend.

Constructing figures from different gels, fields, exposures, and experimental series is discouraged. When this is necessary the component parts of composite images should be indicated by dividing lines clearly demarcated in the figure and described in the legend.