跨物种学习的基本理论是基于不同物种的细胞间有共同的编码机制，只是采用类似信号分子来传输对相同刺激做出的反应。基因本体论解决了跨物种生物数据库对基因产品一致性描述的需求，可以作为跨物种预测的桥梁。人体损伤时间推断是法医学重点问题，对确定死因及伤病关系分析至关重要，其难点在于不能进行人体实验，且动物研究成果又无法直接应用于人体。因此，本项目拟在前期利用大鼠损伤后mRNA时序性表达变化规律，建立数学模型进行损伤时间推断研究的基础上，进一步对大鼠、猪和人不同损伤时间、损伤程度骨骼肌挫伤组织进行高通量转录组测序，通过主成分分析、GO标签和pathway分析建立同源基因数据集，运用种间相关分析、深度信念网络和种间基因集富集分析等算法进行跨物种预测，并通过人体检材的测序数据验证算法的准确性和适用性。最终，构建“鼠-猪-人”损伤时间推断体系，为准确推断人体损伤时间提供科学支持。

The basic theory of cross-species prediction is that there is a common encoding mechanism shared by cells from different species, but distinct signaling molecules are only employed by different species to transmit the signals responding to the same environmental stimuli. Gene ontology addresses the need for consistent description of gene products in cross-species biological databases，so that it can be used as a bridge for cross-species prediction. Dating of wound in humans is an important aspect in forensic medicine. It is vital to determine the cause of death and analyze the relationship between injury and death. The biggest challenge in dating of wound is that not only the incapacity of in vivo experiments, but the inapplicability of animal research results transform to human body directly. Therefore, in this study, on the basis of an established mathematical model that was based on the changes in the temporal expression of mRNA after injury to estimate the wound age in rats, further the high-throughput transcriptome sequencing will be performed on skeletal muscle contusion tissues of rats, pigs and humans with different injury degree and different time periods. Gene data sets will be established through principal component analysis, GO tags and pathway analysis respectively. The interspecific genetic data sets correlation analysis, deep belief network and interspecific gene enrichment algorithms will be performed to do cross-species prediction in dating of wound, then the accuracy and the ability of applicability among these algorithms will be verified by the sequencing data of human physical examination materials. Above all the "rat-pig-human" dating of wound system will be constructed, and as a result, to enhance the accuracy of wound age estimation in human will provide scientific supports.