An epistasis and heterogeneity analysis method based on maximum correlation and maximum consistence criteria

<abstract> <p>Tumor heterogeneity significantly increases the difficulty of tumor treatment. The same drugs and treatment methods have different effects on different tumor subtypes. Therefore, tumor heterogeneity is one of the main sources of poor prognosis, recurrence and metastasis. At present, there have been some computational methods to study tumor heterogeneity from the level of genome, transcriptome, and histology, but these methods still have certain limitations. In this study, we proposed an epistasis and heterogeneity analysis method based on genomic single nucleotide polymorphism (SNP) data. First of all, a maximum correlation and maximum consistence criteria was designed based on Bayesian network score <italic>K2</italic> and information entropy for evaluating genomic epistasis. As the number of SNPs increases, the epistasis combination space increases sharply, resulting in a combination explosion phenomenon. Therefore, we next use an improved genetic algorithm to search the SNP epistatic combination space for identifying potential feasible epistasis solutions. Multiple epistasis solutions represent different pathogenic gene combinations, which may lead to different tumor subtypes, that is, heterogeneity. Finally, the XGBoost classifier is trained with feature SNPs selected that constitute multiple sets of epistatic solutions to verify that considering tumor heterogeneity is beneficial to improve the accuracy of tumor subtype prediction. In order to demonstrate the effectiveness of our method, the power of multiple epistatic recognition and the accuracy of tumor subtype classification measures are evaluated. Extensive simulation results show that our method has better power and prediction accuracy than previous methods.</p> </abstract>

Action-Guided Attention Mining and Relation Reasoning Network for Human-Object Interaction Detection

Human-object interaction (HOI) detection is important to understand human-centric scenes and is challenging due to subtle difference between fine-grained actions, and multiple co-occurring interactions. Most approaches tackle the problems by considering the multi-stream information and even introducing extra knowledge, which suffer from a huge combination space and the non-interactive pair domination problem. In this paper, we propose an Action-Guided attention mining and Relation Reasoning (AGRR) network to solve the problems. Relation reasoning on human-object pairs is performed by exploiting contextual compatibility consistency among pairs to filter out the non-interactive combinations. To better discriminate the subtle difference between fine-grained actions, an action-aware attention based on class activation map is proposed to mine the most relevant features for recognizing HOIs. Extensive experiments on V-COCO and HICO-DET datasets demonstrate the effectiveness of the proposed model compared with the state-of-the-art approaches.

A Method for Analyzing Two-locus Epistasis of Complex Diseases based on Decision Tree and Mutual Entropy

Background: Epistasis makes complex diseases difficult to understand, especially when heterogeneity also exists. Heterogeneity of complex diseases makes the distribution of case population more confused. However, the traditional methods proposed to detect epistasis often ignore heterogeneity, resulting in low power of association studies. Methods: In this study, we firstly use rank information in the Classification Decision Tree and Mutual Entropy (CTME) to construct two different evaluation scores, namely multiple objectives. In addition, we improve the calculation of joint entropy between SNPs and disease label, which elevates the efficiency of CTME. Then, the ant colony algorithm is applied to search two-locus epistatic combination space. To handle the potential heterogeneity, all candidate two-locus SNPs are merged to recognize multiple different epistatic combinations. Finally, all these solutions are tested by χ2 test. Results and Conclusion: Experiments show that our method CTME improves the power of association study. More importantly, CTME also detects multiple epistatic SNPs contributing to heterogeneity. The experimental results show that CTME has advantages on power and efficiency.

The novelty ‘sweet spot’ of invention

Invention arises from novel combinations of prior technologies. However, prior studies of creativity have suggested that overly novel combinations may be harmful to invention. Apart from the factors of expertise, market, etc., there may be such a thing as ‘too much’ or ‘too little’ novelty that will determine an invention’s future value, but little empirical evidence exists in the literature. Using technical patents as the proxy of inventions, our analysis of 3.9 million patents identifies a clear ‘sweet spot’ in which the mix of novel combinations of prior technologies favors an invention’s eventual success. Specifically, we found that the invention categories with the highest mean values and hit rates have moderate novelty in the center of their combination space and high novelty in the extreme of their combination space. Too much or too little central novelty suppresses the positive contribution of extreme novelty in the invention. Furthermore, the combination of scientific and broader knowledge beyond patentable technologies creates additional value for invention and enlarges the advantage of the novelty sweet spot. These findings may further enable data-driven methods both for assessing invention novelty and for profiling inventors, and may inspire a new strand of data-driven design research and practice.