摘要：【中文】本申请公开了一种突变数据识别方法、训练方法、处理装置及存储介质。本申请公开能够同时整合大量非编码区调控因子测序数据和少量非编码区已知有害突变的方法，克服了深度学习模型容易过拟合的缺点，从而对非编码区突变的有害性做出有效的预测。第二，本申请计算效率高，能够在数小时内即能完成模型训练。第三，本申请仅需基序列信息即能完成对非编码区突变有害性的预测，不需要基因表达矩阵等信息的辅助，方便使用。 【EN】The application discloses a mutation data identification method, a training method, a processing device and a storage medium. The application discloses a method capable of simultaneously integrating a large amount of non-coding region regulatory factor sequencing data and a small amount of known harmful mutations of non-coding regions, and overcomes the defect that a deep learning model is easy to overfit, so that the harmfulness of the non-coding region mutations can be effectively predicted. Secondly, the method is high in calculation efficiency, and model training can be completed within hours. Thirdly, the method can predict the harmfulness of the mutation of the non-coding region only by the base sequence information without the assistance of information such as a gene expression matrix and the like, and is convenient to use.

摘要：【中文】本发明公开了一种Hi‑C数据分辨率增强方法及系统，该系统包括：数据输入模块、降采样模块、数据转换模块、生成器模块、判别器模块和数据输出模块。该系统以高分辨率的Hi‑C数据为样本进行训练，计算出低分辨率Hi‑C数据与高分辨率Hi‑C数据之间的映射关系，从而依据该映射关系将分辨率较低的Hi‑C数据增强为分辨率较高的Hi‑C数据。 【EN】The invention discloses a Hi-C data resolution enhancement method and a system, wherein the system comprises the following steps: the device comprises a data input module, a down-sampling module, a data conversion module, a generator module, a discriminator module and a data output module. The system takes Hi-C data with high resolution as a sample for training, and calculates the mapping relation between the Hi-C data with low resolution and the Hi-C data with high resolution, so that the Hi-C data with lower resolution is enhanced into the Hi-C data with higher resolution according to the mapping relation.

摘要：【中文】本发明公开基于区块链的专业知识体系协同构建与审核的方法，其包括以下步骤：步骤1，系统以web服务形式提供专业人员使用，同时知识体系以目录树形式进行组织，目录树中专业人员可以在线协同进行知识节点的创建，并且操作上链；步骤2，其他专业人员对知识节点协同审核，并且操作上链；步骤3，判断知识点是否通过审核；当一个知识节点没有通过审核，则修改后执行步骤2并操作上链；当一个知识节点通过审核，则这个知识节点完成构建，并且操作上链。本发明由专业人员协同构建知识体系并协同审核，不仅提高了构建效率，而且提高了知识体系的准确性和可靠性。进一步，本发明基于区块链的知识体系构建，使得知识体系可溯源，整个过程公开透明提高知识体系的公信力。 【EN】The invention discloses a method for collaborative construction and audit of a professional knowledge system based on a block chain, which comprises the following steps: step 1, the system provides professional personnel for use in a web service form, meanwhile, a knowledge system is organized in a directory tree form, the professional personnel in the directory tree can collaboratively establish knowledge nodes on line, and operate chaining; step 2, performing collaborative audit on the knowledge nodes by other professionals, and operating uplink; step 3, judging whether the knowledge points pass the audit or not; when one knowledge node does not pass the audit, executing the step 2 and operating the uplink after modification; when a knowledge node passes the audit, the knowledge node completes construction and operates uplink. According to the invention, professionals collaboratively construct a knowledge system and collaboratively audit, so that not only is the construction efficiency improved, but also the accuracy and reliability of the knowledge system are improved. Further, the knowledge system is constructed based on the block chain, so that the knowledge system is traceable, and the public and transparent whole process improves the public credibility of the knowledge system.

摘要：【中文】本发明提供一种医疗数据系统的权限管理方法，包括：采取默克尔有向无环图存储医疗数据的权限为记录，记录包括内容、内容哈希值、前左记录哈希值、前右记录哈希值和记录哈希值，内容包括首记录哈希值、受权人、授权记录哈希值、权限、授权日期和有效期，将医疗数据在医疗数据系统生成时的权限对应的记录作为首记录，首记录哈希值是首记录的记录哈希值，授权记录哈希值是授权人获得授权权利的记录的记录哈希值，记录哈希值是本记录的内容哈希值、前左记录哈希值和前右记录哈希值三者拼接后，通过哈希运算生成的哈希值。上述方法能够快速检验权限历史篡改情况。 【EN】The invention provides a method for managing the authority of a medical data system, which comprises the following steps: the method comprises the steps of adopting the authority of storing medical data by a Mercker directed acyclic graph as a record, wherein the record comprises content, a content hash value, a front left record hash value, a front right record hash value and a record hash value, the content comprises a first record hash value, an authorized person, an authorized record hash value, an authority, an authorization date and an effective period, taking the record corresponding to the authority of the medical data generated in a medical data system as a first record, the first record hash value is the record hash value of the first record, the authorized record hash value is the record hash value of the record for which the authorized person obtains the authorization right, and the record hash value is the hash value generated by hash operation after the content hash value, the front left record hash value and the front right record hash value of the record are spliced. The method can quickly check the historical tampering condition of the authority.

摘要：【中文】本发明提供一种基于边缘计算的医疗影像识别方法及系统，包括：数据使用端向域内边缘计算发出第一请求，第一请求包括算法、数据请求信息和私钥；边缘计算根据第一请求在云中心查找符合第一请求的数据拥有端，建立沙盒，使用私钥对算法进行加密，加密后算法和数据请求信息放进沙盒并发送到云中心；云中心将数据请求信息发送到符合第一请求的数据拥有端；数据拥有端查询数据请求信息，将查询结果对应的数据集用私钥加密后发送到云中心；云中心对沙盒中的算法和数据集进行解密，对数据集执行算法，获得图像识别结果集，进行加密，发送给数据拥有端和数据使用端。上述方法和系统提供可靠算法运行方法，确保医疗影像数据不会从数据源泄露到外部。 【EN】The invention provides a medical image identification method and system based on edge calculation, which comprises the following steps: the data using end sends a first request to the inner edge calculation, wherein the first request comprises an algorithm, data request information and a private key; the edge computing searches a data owning end which accords with the first request in the cloud center according to the first request, establishes a sandbox, encrypts an algorithm by using a private key, and places the encrypted algorithm and data request information into the sandbox and sends the encrypted algorithm and data request information to the cloud center; the cloud center sends the data request information to a data owning end meeting the first request; the data owner inquires data request information, encrypts a data set corresponding to an inquiry result by using a private key and then sends the encrypted data set to the cloud center; and the cloud center decrypts the algorithm and the data set in the sandbox, executes the algorithm on the data set to obtain an image identification result set, encrypts the image identification result set and sends the encrypted image identification result set to the data owning end and the data using end. The method and system provide a reliable algorithm operation method to ensure that medical image data is not leaked from a data source to the outside.