Sequence processing method and apparatus, posterior matrix processing method and apparatus, and computing device

An apparatus for determining a polymer unit sequence of a polymer, the apparatus comprising:

• a processing module configured to process a raw signal recorded when performing a nanopore sequencing on the polymer at a series of time steps to obtain a first vectorized representation sequence carrying an increased information payload than the raw signal;
• a weight acquisition module coupled to the processing module and configured to input the first vectorized representation sequence into a first machine learning model to obtain a weight vector output, wherein the weight vector is as long as the first vectorized representation sequence and comprises a plurality of predicted translocation values, each predicted translocation value indicating how many polymer units the polymer has moved relative to an observation site of the nanopore sequencing at a corresponding time step;
• an integration module coupled to the weight acquisition module and configured to integrate the first vectorized representation sequence with the weight vector to obtain a second vectorized representation sequence characterizing the polymer;
• and
• a determination module coupled to the integration module and configured to determine the polymer unit sequence of the polymer with improved accuracy by virtue of the increased information payload based on, at least in part, the second vectorized representation sequence.

A computer-readable storage medium having one or more code instructions stored thereon, wherein the one or more code instructions, when executed by one or more processors, cause the one or more processors to perform operations comprising:

• processing a raw signal recorded when performing a nanopore sequencing on a polymer at a series of time steps to obtain a first vectorized representation sequence carrying an increased information payload than the raw signal;
• inputting the first vectorized representation sequence into a first machine learning model to obtain a weight vector, wherein the weight vector is as long as the first vectorized representation sequence and comprises a plurality of predicted translocation values, each predicted translocation value indicating how many polymer units the polymer has moved relative to an observation site of the nanopore sequencing at a corresponding time step;
• integrating the first vectorized representation sequence with the weight vector to obtain a second vectorized representation sequence characterizing the polymer;
• and
• determining a polymer unit sequence of the polymer with improved accuracy by virtue of the increased information payload based on, at least in part, the second vectorized representation sequence.