In Imaging Transcriptomics methods, such as MERFISH, STARmap, and SEQfish, samples are passed through multiple rounds of multi-color imaging with single-molecule resolution, and the sequence of colors originating from individual molecules is used to assign each molecule to a gene identity, encoded by a molecularly designed codebook. This is done with single molecule resolution, and thus analyzing an organ like an entire mouse brain with IT would require years of instrument time. A human brain atlas would require thousands of times longer. We approach this problem by considering the structure of the data and finding inefficiencies in the sampling strategy. We then develop compressed sensing and machine learning methods to recover the same information from more efficiently sampled data. Finally, we implement these methods in the form of newly designed optical systems and biochemical strategies. This work is performed in close collaboration with Dr. Brian Cleary at the Broad and Dr. Yonina Eldar at the Weizmann Institute.