Genemind Biosciences Company Limited

Tencent Technology (shenzhen) Company Limited

Direct Genomics Co., Ltd.

Huan Jin

Zefei Jiang

Qin Yan

Zhiliang Zhou

Weibin Xu

Linsen Li

Hongbo Jin

Ye Tao

Xiong Li

Yin Yan Li

Anqun Pan

Songjian Wang

Jianwei Deng

Cheng Yang

Fang Chen

Ruizhou Wu

Lei Sun

Wutong Jiang

Dongpo Bao

Jicai Fan

Zeifei Jiang

The present disclosure includes a method for training a sequence mining model. In the method, a first sequence sample in a target service scenario is obtained. A tag status of the first sequence sample is obtained, the tag status of the first sequence sample indicating a proportion of the first sequence sample that has corresponding tag information. A sub-model from a sequence mining frame is selected according to the tag status to construct the sequence mining model. Also, the sequence mining model is trained by using the first sequence sample. The sequence mining frame includes a first sub-model configured to obtain a latent representation, a second sub-model configured to determine the target tag information when the tag status meets a first condition, and a third sub-model being configured to determine the target tag information when the tag status meets a second condition.

Method for training sequence mining model, method for processing sequence data, and device

A method for identifying a base in nucleic acid, a computer-readable storage medium, a computer program product, and a system. The method for identifying a base in nucleic acid comprises: mapping a coordinate of each bright spot in a bright spot set corresponding to a template onto an image to be inspected, and determining the position of a corresponding coordinate on said image (S); determining the intensity of a signal at the position of the corresponding coordinate on said image, the intensity being a corrected intensity (S); and comparing the intensity of the signal at the position of the corresponding coordinate on said image with the size of a first preset value, and determining a base type corresponding to the position on the basis of the comparison result, so as to achieve base calling (S). The method may quickly and accurately identify a base, and achieve the determination of an order of nucleotides/bases of at least part of a sequence of a template.

Method for identifying base in nucleic acid and system

A method and device for image registration, and a computer program product. The method includes performing, based on a reference image, a first registration of an image to be registered, including determining a first offset between predetermined areas on the image and the reference image, and moving all bright spots on the image based on the first offset; performing, based on the reference image, a second registration with respect to the first registered image, including merging the first registered image and the reference image, calculating the offset of all overlapped bright spots in the predetermined area on the merged image to determine a second offset, two or more bright spots distanced by less than a predetermined number of pixels being one overlapped bright spot, and moving, on the basis of the second offset, all bright spots on the first registered image, thus implementing the registration of the image.

Method and device for image registration, and computer program product

A method and device for detecting bright spots on an image. The image is acquired from a field of view where a base extension reaction occurs, multiple nucleic acid molecules with optically detectable labels exist in the field of view where the base extension reaction occurs, and at least some of the nucleic acid molecules appear as bright spots on the image. The method includes preprocessing the image to obtain a preprocessed image; determining a critical value to simplify the preprocessed image to obtain a simplified image, determining a first bright spot detection threshold on the basis of the preprocessed image, and identifying candidate bright spots on the image on the basis of the preprocessed image and the simplified image, including determining a pixel matrix that satisfies at least two of conditions as a candidate bright spot.

Method and device for detecting bright spots on image, and computer program product

The disclosure discloses an alignment method, device, and system. The alignment method includes: converting each read into a set of short fragments corresponding to the read to obtain a plurality of sets of short fragments; determining a corresponding position of the short fragment in a reference library to obtain a first positioning result, wherein the reference library is a hash table constructed based on a reference sequence, the reference library includes a plurality of entries, one entry of the reference library corresponds to one seed sequence, and the seed sequence is capable of matching at least one sequence on the reference sequence, a distance between two seed sequences corresponding to two adjacent entries of the reference library on the reference sequence is less than a length of the short fragment; removing a short fragment positioned on any one of the adjacent entries of the reference library in the first positioning result to obtain a second positioning result; and extending based on short fragments from the same read in the second positioning result to obtain an alignment result of the read. The alignment method can efficiently and accurately process and position sequencing data.

Alignment methods, devices and systems

A method for constructing a sequencing template based on an image, a device, and a system. The image includes first, second, third and fourth images of one same field of view corresponding to base extensions of A, T/U, G, and C respectively; the first, second, third and fourth images respectively include images Mand M, images Nand N, images Pand P, and images Qand Q; the method includes combining any two of the images M, M, N, N, P, P, Q, and Qto perform bright spot matching, and enabling such images to participate in the combination for at least one time to obtain a plurality of combined images including first coincident bright spots, and merging the first coincident bright spots on the plurality of combined images to obtain a bright spot set corresponding to the sequencing template.

Method for constructing sequencing template based on image, and base recognition method and device

Aspects of the disclosure provide a data processing apparatus having interface circuitry and processing circuitry. The interface circuitry is configured to obtain data related to at least two resource accounts, and related to resource transfer of the at least two resource account. The processing circuitry configured to construct resource transfer relationships of the at least two resource accounts according to the data, calculate, based on the resource transfer relationships, similarity measurements between the at least two resource accounts according to a preset calculation rule, and output, according to the calculated similarity measurements, a resource account with a similarity measurement satisfying a preset condition.

Data processing method and apparatus for virtual resource

An identity authentication method is provided, including: obtaining a virtual-resource data processing request sent by a mobile terminal, the virtual-resource data processing request carrying a prestored digital fingerprint, and the digital fingerprint being generated by using device information and user information; performing virtual-resource data processing authentication on the virtual-resource data processing request according to a pre-established user value transfer behavior model; after virtual-resource data processing authentication succeeds, generating a value transfer token according to the digital fingerprint, and returning the value transfer token to the mobile terminal; receiving a value transfer request sent by the mobile terminal, the value transfer request carrying the value transfer token; and checking whether the value transfer token is valid, if the value transfer token is valid, identity authentication succeeding; otherwise, identity authentication failing.

Identity authentication method, server, and storage medium

The present disclosure provides a method and a system for processing an image. The method includes: an image pre-processing step, including analyzing an image input to be processed so as to obtain a first image; and a spot detecting step, including analyzing the first image so as to compute a spot determining threshold; analyzing the first image so as to acquire a candidate pixel spot, and judging whether the candidate pixel spot is the spot according to the spot determining threshold; computing a sub-pixel center coordinate of the spot and an intensity value of the sub-pixel center coordinate if yes; and discarding the candidate pixel spot if no. According to the method, the image is denoised in the image pre-processing step, such that a calculation in the spot detecting step may be reduced. In addition, the accuracy of determining the image spot may be improved.

Method and system for processing an image comprising spots in nucleic acid sequencing

A method and a device for identifying a single molecule, a method and a device for counting a single molecule and a system for processing a single molecule are provided. The method for identifying the single molecule includes inputting time sequence of an intensity of an spot; forming a line chart of time and intensity of the spot according to the time sequence, wherein the line chart consists of a plurality of line segments; dividing the line chart into a plurality of grids in an array, counting the number of the line segments and/or ends of the line segments in each grid; subjecting the divided line chart to a line erosion according to numbers corresponding to the grids to transform the divided line chart to a simplified image; subjecting the simplified image to run-length coding to indicate connected domains; calculating an area of each connected domain and determining that the connected domain corresponds to a single molecule if the area of the connected domain is greater than a first predetermined threshold. According to the method for identifying the single molecule, the line chart of the time sequence of the intensity of the spot is processed to transform into the image, which results in a quick identification for the single molecule and a high accuracy of the identification.

Growing community of inventors

Shenzhen, China

Huan Jin