No abstract

Transcription factors (TFs) are crucial components of regulatory networks that control gene transcription. Current TF assays are limited to the analysis of a single TF or require TF-specific antibodies. Here we report the Single Primer Amplification assisted Oligonucleotide Array-based Transcription Factor Assay (SPA-OATFA) which can directly analyze the binding activities of 240 human TFs simultaneously. Examining early events during serum-stimulation of HeLa cells as a model, we demonstrated the utility of SPA-OATFA combined with whole genome gene expression to systematically map the temporal activation of signaling pathways. Both TFs known to function in this stimulation response such as EGR1 and AP1 and new TFs such as HSF1 were identified. This information, combined with mRNA profiling, provided novel insights into the activities of regulatory pathways, and illustrates the potential of SPA-OATFA in detailed systems biology analysis of cell responses.

Cell/tissue-specific gene expression are tightly regulated by various combinations of multiple transcription factors (TFs). Here, we present an oligonucleotide array-based transcription factor assay (OATFA), which allows the simultaneous assay of multiple active TFs. In this proof-of-principle work, both purified TFs and cell extracts were analyzed using OATFA and further antibody-based validation confirmed the chip data. This method could simplify the assay of multiple TFs and may facilitate high-throughput profiling of large numbers of TFs.

Molecular systematics involves the description of the regulatory networks formed by the interconnections between active transcription factors and their target expressed genes. Here, we have determined the activities of 200 different transcription factors in six mouse tissues using an advanced mouse oligonucleotide array-based transcription factor assay (MOUSE OATFA). The transcription factor signatures from MOUSE OATFA were combined with public mRNA expression profiles to construct experimental transcriptional regulatory networks in each tissue. SRF-centered regulatory networks constructed for lung and skeletal muscle with OATFA data were confirmed by ChIP assays, and revealed examples of novel networks of expressed genes coregulated by sets of transcription factors. The combination of MOUSE OATFA with bioinformatics analysis of expressed genes provides a new paradigm for the comprehensive prediction of the transcriptional systems and their regulatory pathways in mouse.

The precious rare edible fungus Morchella conica is popular worldwide for its rich nutrition, savory flavor, and varieties of bioactive components. Due to its high commercial, nutritional, and medicinal value, it has always been a hot spot. However, the molecular mechanism and endophytic bacterial communities in M. conica were poorly understood. In this study, we sequenced, assembled, and analyzed the genome of M. conica SH. Transcriptome analysis reveals significant differences between the mycelia and fruiting body. As shown in this study, 1,329 and 2,796 genes were specifically expressed in the mycelia and fruiting body, respectively. The Gene Ontology (GO) enrichment showed that RNA polymerase II transcription activity-related genes were enriched in the mycelium-specific gene cluster, and nucleotide binding-related genes were enriched in the fruiting body-specific gene cluster. Further analysis of differentially expressed genes in different development stages resulted in finding two groups with distinct expression patterns. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment displays that glycan degradation and ABC transporters were enriched in the group 1 with low expressed level in the mycelia, while taurine and hypotaurine metabolismand tyrosine metabolism-related genes were significantly enriched in the group 2 with high expressed level in mycelia. Moreover, a dynamic shift of bacterial communities in the developing fruiting body was detected by 16S rRNA sequencing, and co-expression analysis suggested that bacterial communities might play an important role in regulating gene expression. Taken together, our study provided a better understanding of the molecular biology of M. conica SH and direction for future research on artificial cultivation.

No abstract

Cold-start problem has been recognized as the most crucial challenge in recommender systems. Many recommendation algorithms work well when lots of preference information is available but start to degrade in cold-start settings. Inspired by the spirit of meta-learning, we identified that the appeal of cold-start problems and the superiority of meta-learning are congruous where meta-learning aims to learn a model from a small set of labeled examples (users' consuming history), and then this model can be quickly generalized to new tasks (recommendation for new users or new items). Therefore, faced with the extreme cold-start scenario, we proposed a meta-learning embedding ensemble (ML2E) recommendation algorithm to forecast new users' preference and generate desirable initial embedding for new items. It's worth pointing out that the training process of ML2E only uses few first-order gradient information, and ML2E not only has the incremental mining ability for different mini-batches on the same task but also the generalization ability for different tasks. Finally, we validated ML2E on two benchmark datasets, experimental results showed that our algorithm has significantly improved recommendation metric in comparison with three existing baselines. INDEX TERMS Cold-start setting, meta learning-based embedding, online recommendation algorithms.

No abstract