Various systems for the production of useful proteins have been developed using the fission yeast Schizosaccharomyces pombe as a host, and some are now being used commercially. It is necessary, however, to improve the system further for the production of low‐cost chemicals and commodities, so that the host becomes more economical and productive and can be widely used for the production of different molecules. We hypothesized that many S. pombe genes are not necessary under nutrient‐rich growth conditions; or rather, they serve only to waste energy when seen from the viewpoint of protein production, because their products are necessary only for adaptation to different environments. Thus we have tried to create S. pombe mutants that are dedicated to heterologous protein production by deleting as many non‐essential genes as possible. Putative essential genes were mapped using the genome information of S. pombe. The transcriptome of gene disruptants was analysed using microarrays and, using this system, a new promoter was identified. The method (called the Latour method) has been developed to delete efficiently a large region from the chromosome, resulting in the establishment of mutant strains lacking approx. 500 kb of genetic material. New experimental strains auxotrophic for six nutrients were established that were conveniently used for co‐expression of proteins using multiple plasmids. An efficient transformation method has also been developed that is useful for investigating heterologous protein production in a variety of strains. Incidentally, in heterologous protein production systems, products are often degraded, leading to a decline in production efficiency. Thus, to examine heterologous protein production, we created 52 S. pombe mutant strains in each of which a single protease gene was destroyed. We also successfully constructed strains in which multiple protease genes were disrupted. As a result, it was shown that the production of a model protein, human growth hormone, was increased in this strain. Furthermore, we obtained many strains that lacked genes related to glucose metabolism, intracellular transport or biosynthesis of sugar chains. The present minireview covers the results of functional analysis of these strains. By preparing strains in which large chromosomal regions have been deleted and then combining strains defective in various functional genes, the establishment of effective hosts will become possible.