||Image-based sequencing of mRNA makes it possible to see where in a tissue sample a given gene is active, and thus discern large numbers of different cell types in parallel. This is crucial for gaining a better understanding of tissue development and disease such as cancer. To achieve this goal, signals of targeted mRNA are collected over multiple staining and imaging cycles, but signal density together with noise makes signal decoding challenging.
Previous approaches have led to low signal recall in efforts to maintain high sensitivity.
I will present a new approach in which recall has been increased, at maintained sensitivity, thanks to: a different way of selecting signal candidates, the introduction of a simple CNN to determine how much candidates look like real signals, and multiple strategies of decoding sequences depending on dataset complexity.
Joint work with Gabriele Partel and Carolina Wählby