Hardware acceleration method for predication of RNA second-stage structure with pseudoknot

Abstract

The invention discloses a method for accelerating the predication of an RNA second-stage structure with pseudoknot based on a four-dimensional dynamic planning method, and aims at accelerating the predication of the RNA second-stage structure with pseudoknot. According to the technical scheme, the method comprises the steps of building a heterogeneous computing system through a host and a reconfigurable algorithm accelerator; sending parameters of a formatted thermodynamic model and coded RNA sequences to the reconfigurable algorithm accelerator through the host; computing seven computing modules of the algorithm accelerator through the non-backtracking PKNOTS algorithm by the MPMD mode; when in computing, a four-dimensional matrix is decomposed by the rectangular dimension reduction method into N three-dimensional matrixes, then the fine granularity is achieved by the task dividing strategy of circularly dividing in each layer by areas and parallelly processing by rows in the area, and the computing is carried out synchronously; for n PE in each computing module, n data in different rows of the area are computed synchronously through the SPMD mode. With the adoption of the method, the predication of the RNA second-stage structure with pseudoknot is accelerated; the technology is novel, the performance is high, and the cost is low.

Description

technical field [0001] The invention relates to a method for accelerating the prediction of the RNA secondary structure with pseudoknots based on a four-dimensional dynamic programming method, and aims to speed up the prediction speed of the RNA secondary structure with pseudoknots. Background technique [0002] The secondary structure of RNA is an important basis for identifying ncRNA, and it is the basis and premise of studying RNA function. Experimental methods are the most reliable method to obtain the secondary structure of RNA. At present, the main methods for determining RNA structure are X-ray diffraction and nuclear magnetic resonance. Although the results obtained by experimental methods are accurate and reliable, the process is very time-consuming and expensive. Therefore, research The calculation method of RNA structure prediction is particularly important. In recent years, the method of predicting the secondary structure of RNA sequence using computer and mathem...

AI Technical Summary

Problems solved by technology

At present, although there are many improved algorithms based on PKNOTS that can support false knot prediction, these algorithms have sacrificed accuracy and correctness for execution speed, and the prediction effect is not good; Weight matching algorithms, while having ideal computational complexity, only have good predictive results for specific types of false knots

In comparison, the prediction effect of the PKNOTS algorithm is significantly better than other algorithms, but the high space-time complexity limits the practicability of the PKNOTS algorithm. Currently, it can only predict short sequence structures containing dozens of bases

[0022] In 2010, based on the IBM Cell multi-core processor, Krishnan et al. conducted the first parallel research on the four-dimensional dynamic programming algorithm based on the PKNOTS algorithm. Compared with the standard PKNOTS software running on the general-purpose microprocessor, the Sony Play Station3 The parallel version running on the platform has obtained about 3 times the acceleration effect, but it can only support RNA sequences with a length less than 100bps

[0032] The survey results show that most of the current research on hardware acceleration in the field of sequence analysis at home and abroad is limited to the first-level structure of the sequence, and there is currently no report on hardware acceleration for high-dimensional, especially four-dimensional structure prediction algorithms.

Images