Search

Additional moisture in preparing red koji, Monascus-fermented rice, is a characteristic production process. To determine how additional moisture affects red koji preparation as per quality, we compared the growth of Monascus purpureus, enzyme and pigment production, and related gene expressions using our findings. We considered two kinds of red koji: one prepared with additional moisture at the middle part of the preparation and the other prepared without additional moisture. Our results showed that additional moisture did not promote the growth of M. purpureus, but it was significantly increased the pigment (red and yellow) and tended to increase the α-amylase level and saccharification power. Although adding a high amount of moisture (approximately 60% moisture content) promoted pigment production, it slightly repressed enzyme production. In contrast, adding approximately 50% moisture content promoted enzyme production. These findings showed that the additional moisture can affect the quality of red koji on the purpose. The expression of 10 pigment biosynthetic gene clusters and two glycohydrolase genes in red koji after adding moisture was analyzed through real-time qPCR. Eight genes were upregulated within 1 hr after adding water, with mppR2 being the first upregulated gene within 30 min. The expression of genes as per pigment production quickly responded to additional moisture during solid-state fermentation. Moreover, acetyl-CoA, which is a starting substrate for pigment content in red koji was increased within 3 hr after adding water. This study first described the relationship between additional moisture and expression of pigment biosynthetic genes by Monascus spp. during red koji preparation.

When facing rich multimedia content and making a decision, users tend to be overwhelmed with redundant options. Recommendation system can improve the users' experience by predicting the possible preference of a given user. The vast majority of the literature adopts the collaborative framework, which relies on a static and fixed formulation of the rating score prediction function (in most cases an inner product function). However, such a static learning paradigm is not consistent with the dynamic feature of human intelligence. Motivated by this, we present a novel adversarial framework for collaborative ranking. On one hand, we leverage a deep generator to approximate an arbitrary continuous score function in terms of pairwise comparison. On the other hand, a discriminator provides personalized supervision signals with increasing difficulty. Different from the traditional static learning framework, our proposed approach enjoys a dynamic nature and unifies both the generative and the discriminative model for collaborative ranking. Comprehensive empirical studies on three real-world datasets show significant improvements of the adversarial framework over the state-of-the-art methods.