In-Memory mRNA Quantification via Bitwise k-mer Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for mRNA quantification in bioinformatics are time-consuming and require significant computational resources, posing challenges for efficient and accurate analysis in fields like disease prediction and drug development.
Innovation Solution
A Processing-In-Memory (PIM) architecture and algorithm that stores an index table of nucleotide substrings in non-volatile memory, allowing for parallel computing and reducing data movement, enabling fast and efficient mRNA quantification by using bitwise operations such as XNOR and AND to match short reads with the index table, thereby calculating the abundance of mRNA sequences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional alignment-based methods are used for mRNA quantification, then mapping accuracy is improved, but processing time and computational resources increase significantly
Solution Approach 1:
The patent segments the reference genome into k-mer substrings (short sequences of length k) and stores them in a hash table with their corresponding genomic positions. This segmentation allows the system to compare only relevant k-mers from the input reads against the pre-processed index, rather than performing full read-to-genome alignment. The segmentation principle resolves the contradiction by maintaining mapping accuracy through precise k-mer matching while dramatically reducing processing time through efficient hash table lookups.
Solution Approach 2:
The patent performs preliminary action by pre-processing the reference genome into k-mer fragments and storing them in a hash table before actual quantification occurs. This pre-computation includes extracting all possible k-mers, recording their positions, and organizing them for rapid access. During quantification, only hash table lookups are needed rather than full alignment procedures. This preliminary preparation resolves the time-accuracy contradiction by shifting computational work to an offline phase, enabling fast online queries that maintain mapping precision.
2Measurement precision
If full base-to-base alignment of reads is performed, then mapping precision is improved, but alignment efficiency decreases
Solution Approach 1:
The patent extracts only the essential k-mer substrings from the full read sequences and compares these extracted portions against the k-mer index. Instead of performing computationally intensive base-to-base alignment of entire reads, the system extracts and compares only the critical k-mer portions that are sufficient for accurate mapping. This extraction principle resolves the contradiction by maintaining mapping precision through focused k-mer comparison while dramatically improving alignment efficiency by avoiding unnecessary full-read processing.
Solution Approach 2:
The patent creates a copied and simplified representation of the genome data in the form of a hash table containing only k-mer sequences and their positions, rather than working with the complete genomic sequence. This copied structure enables rapid lookup operations during quantification. The copying principle resolves the efficiency-precision contradiction by providing a streamlined data representation that maintains all necessary mapping information while enabling much faster comparison operations.
3Productivity
If alignment-free k-mer based counting algorithms are used, then quantification efficiency is improved, but computational resources are still significantly required
Solution Approach 1:
The patent merges the indexing and counting functions into a unified hash table structure that stores k-mers along with their frequency counts and positional information. Instead of separate operations for indexing and quantification, the system combines these functions, allowing simultaneous update of counts during the lookup process. This merging principle resolves the contradiction by improving quantification efficiency through integrated operations while reducing computational resource requirements by eliminating redundant processing steps.
Data Source
AI summary
A method of calculating an abundance of an mRNA sequence within a gene comprises storing an index table of the gene in a non-volatile memory, obtaining a short read of the mRNA sequence, generating a set of input fragments from the mRNA sequence, initializing a compatibility table in a volatile memory, for each input fragment in the set of input fragments, searching for an exact match of the input fragment in the index table, calculating a final result from the compatibility table, and calculating an abundance of the mRNA sequence in the gene by aggregating the transcripts compatible with the short read, wherein the calculating step is performed on the same integrated circuit as the non-volatile memory. A system for in-memory calculation of an abundance of an mRNA sequence within a gene is also disclosed.