Dhr96[1] mutation and maternal tudor[1] mutation increase life span and reduce the beneficial effects of mifepristone in mated female Drosophila.

Mating and receipt of male Sex Peptide hormone cause increased egg laying, increased midgut size and decreased life span in female Drosophila. Feeding mated females with the synthetic steroid mifepristone decreases egg production, reduces midgut size, and increases life span. Here, several gene mutations were assayed to investigate possible mechanisms for mifepristone action. Drosophila Dhr96 is a hormone receptor, and a key positive regulator of midgut lipid uptake and metabolism. Dhr96[1] null mutation increased female life span, and reduced the effects of mifepristone on life span, suggesting that Dhr96[1] mutation and mifepristone may act in part through the same mechanism. Consistent with this idea, lipidomics analysis revealed that mating increases whole-body levels of triglycerides and fatty-acids in triglycerides, and these changes are reversed by mifepristone. Maternal tudor[1] mutation results in females that lack the germ-line and produce no eggs. Maternal tudor[1] mutation increased mated female life span, and reduced but did not eliminate the effects of mating and mifepristone on life span. This indicates that decreased egg production may be related to the life span benefits of mifepristone, but is not essential. Mifepristone increases life span in w[1118] mutant mated females, but did not increase life span in w[1118] mutant virgin females. Mifepristone decreased egg production in w[1118] mutant virgin females, indicating that decreased egg production is not sufficient for mifepristone to increase life span. Mifepristone increases life span in virgin females of some, but not all, white[+] and mini-white[+] strains. Backcrossing of mini-white[+] transgenes into the w[1118] background was not sufficient to confer a life span response to mifepristone in virgin females. Taken together, the data support the hypothesis that mechanisms for mifepristone life span increase involve reduced lipid uptake and/or metabolism, and suggest that mifepristone may increase life span in mated females and virgin females through partly different mechanisms.

A graph placement methodology for fast chip design.

Chip floorplanning is the engineering task of designing the physical layout of a computer chip. Despite five decades of research1, chip floorplanning has defied automation, requiring months of intense effort by physical design engineers to produce manufacturable layouts. Here we present a deep reinforcement learning approach to chip floorplanning. In under six hours, our method automatically generates chip floorplans that are superior or comparable to those produced by humans in all key metrics, including power consumption, performance and chip area. To achieve this, we pose chip floorplanning as a reinforcement learning problem, and develop an edge-based graph convolutional neural network architecture capable of learning rich and transferable representations of the chip. As a result, our method utilizes past experience to become better and faster at solving new instances of the problem, allowing chip design to be performed by artificial agents with more experience than any human designer. Our method was used to design the next generation of Google's artificial intelligence (AI) accelerators, and has the potential to save thousands of hours of human effort for each new generation. Finally, we believe that more powerful AI-designed hardware will fuel advances in AI, creating a symbiotic relationship between the two fields.