Since the microscope is the primary instrument for data acquisition in spatial biology, it is useful to understand its parts and functionality to ensure the best data quality and to troubleshoot when there are issues with acquisition.

Fluorescence microscopes

Bright field microscopes

Troubleshooting

Blurry images

Blurry images are caused by two main things:

• Image is out of focus
• Debris in the optical path

In both cases, the image will need to be reacquired, and the focal plane should be adjusted until the objects are in focus. Debris should be carefully removed from the optical path, either by gentle washing if it’s on the sample or with lens wipes if the debris is on the microscope.

Grainy images

In order for an object to be detected clearly, the real signal must be far enough above the noise (high signal-to-noise ratio) that the noise can be easily taken out through thresholding. Grainy images are characteristic of having low signal-to-noise ratio (SNR).

To improve noisy image quality, there are two options: increase signal strength or decrease noise. However, it is generally easier to increase signal strength because most noise sources are dictated by the physics of the microscope/camera/computer system.

Increasing signal strength

• Increase fluorescent dye concentration: This will only work up to the point of saturation of the dye molecule. For example, if the target of interest is a protein and the dye is conjugated to the protein’s antibody, then at a certain point, no more antibody can bind to the target protein. The low signal is therefore a result of low expression rather than an imaging issue.
• Increase the exposure time: By increasing the exposure time, this increases the time that the fluorescent molecule is exposed to the excitation light, which will allow it to go through more excitation-emission cycles, thus increasing the number of photons that are emitted. This also increases the time that the camera sensor collects photons. However, higher exposure times can lead to photobleaching, where the fluorescent molecule is no longer able to change conformations and thus emit photons.
• Increase gain: The camera gain determines how much signal amplification occurs between the photon hitting the sensor and the sensor converting it into a voltage. This generally does not change the SNR very much, but can make it possible to get better separation between the signal and noise such that the noise can be thresholded out.
• Increase light intensity: Higher light source power increases the number of photons that are output by the light source, which increases the chance that the fluorophores are excited. However, this also has the effect of increasing the rate of photobleaching. Furthermore, it can also lead to higher temperatures on the sample, which may lead to sample degradation.
• Binning: Camera binning combines the information from neighboring pixels, and generally increases the signal strength in comparison to readout noise. However, this comes at the expense of resolution. Note that if the data is binned after acquisition, the SNR will remain the same, but similar to gain, it may help with thresholding.

Decreasing noise

• Keeping the microscope room cool: Because dark current is a consequence of thermal noise, it is generally improved with lower temperatures. Most cameras used for recording fluorescent signal are integrated with some sort of cooling system to keep the temperature of the sensor chip low. However, if the sensor overheats, the noise will be very high. Generally overheating will fix itself if the camera is turned off and given time to cool down again. Maintaining the microscope room at a cooler temperature does help the camera, but the best thing to do is to give the camera time to cool off if overheating is a problem.