Systems and methods for SNP analysis and genome sequencing

Inventors

Thomas, Sterling • Dellinger, Nathan

Assignees

Noblis Inc

Abstract

In some embodiments, techniques for identifying one or more species in an undifferentiated environmental sample comprising a plurality of nucleic acid sequences are provided. One or more indices that represent a plurality of reference nucleic acid sequences may be provided, and data may be received comprising digital representations of respective nucleic acid sequences. The respective nucleic acid sequences may be aligned, if possible, using the indices. A respective alignment ration may be calculated for each one of the reference nucleic acid sequences, based on the number of nucleic acid sequences aligned to the respective reference nucleic acid sequence and the total number of nucleic acid sequences in the received data.

Core Innovation

The invention relates to identifying one or more species in an undifferentiated environmental sample using a computer system that aligns received nucleic acid sequences to reference nucleic acid sequences. Reference subsequences are retrieved from positions of reference nucleic acid sequences associated with generated indices, and a hash is computed to determine a corresponding element of the generated index. Position data is stored in the index element reflecting the first position and indicating the reference nucleic acid sequence associated with the generated index.

The system loads one or more indices representing a plurality of reference nucleic acid sequences, where each index comprises elements corresponding to potential nucleic acid sequence permutations. The one or more indices are limited, based on statistical methods regarding which permutations are most likely to occur, to less than a total possible number of permutations. The system receives data comprising data structures digitally representing respective nucleic acid sequences, and for each nucleic acid sequence attempts to align to each reference nucleic acid sequence using the one or more indices.

For each received nucleic acid sequence, the system determines using the one or more indices whether the nucleic acid sequence is successfully aligned to any one or more reference nucleic acid sequences. For each reference nucleic acid sequence, the system calculates a respective alignment ratio based on the number of nucleic acid sequences determined to be successfully aligned to the respective reference nucleic acid sequence and based on the total number of nucleic acid sequences in the received data. The system generates and stores the plurality of calculated alignment ratios for the received data.

Claims Coverage

The identified independent claims are directed to a system, a non-transitory computer-readable storage medium, and a method, each covering the same core approach with generated hash-based indices, indexed alignment attempts, and per-reference alignment ratio calculation. Across the independent claims, the main inventive features number 8.

Across independent claims, the coverage is centered on a processor/memory-based approach that generates and stores hash-based indices that map reference subsequences from positions to index elements containing position data, limits the index contents to a statistically likely set of permutations, uses the indices to attempt alignments for each received nucleic acid sequence and determine alignment success, and computes and stores a per-reference alignment ratio derived from counts of successfully aligned sequences relative to the total received sequences.

Stated Advantages

Documented Applications