Bioinformatics systems, apparatuses, and methods executed on an integrated circuit processing platform
A system, method and apparatus for executing a bioinformatics analysis on genetic sequence data includes an integrated circuit formed of a set of hardwired digital logic circuits that are interconnected by physical electrical interconnects. One of the physical electrical interconnects forms an input to the integrated circuit that may be connected with an electronic data source for receiving reads of genomic data. The hardwired digital logic circuits may be arranged as a set of processing engines, each processing engine being formed of a subset of the hardwired digital logic circuits to perform one or more steps in the bioinformatics analysis on the reads of genomic data. Each subset of the hardwired digital logic circuits may be formed in a wired configuration to perform the one or more steps in the bioinformatics analysis.
1. A genomics analysis platform for executing a sequence analysis pipeline, the genomics analysis platform comprising:
one or more first integrated circuits, each first integrated circuit forming a processing unit that is responsive to one or more software algorithms that are configured to instruct the processing unit to perform one or more of genomic processing steps of the sequence analysis pipeline, the processing unit having a first set of physical electronic interconnects;
one or more second integrated circuits, each second integrated circuit forming a programmable logic device, the programmable logic device having one or more hardware logic circuits that are programmatically configured to arrange the hardwired digital logic circuits to perform one or more different genomic processing steps of the sequence analysis pipeline, the hardwired digital logic circuits of each programmable logic device being arranged as a set of processing engines to perform the one or more different genomic processing steps; and
a shared memory electronically connected to the one or more first integrated circuits via at least a portion of the first set of physical electronic interconnects, the shared memory storing genetic sequence data, wherein each of the one or more first integrated circuits and each of the one or more second integrated circuits are configured to access the genetic sequence data in the shared memory based on one or more cache coherency protocols.
2. The genomics analysis platform of claim 1 , wherein the cache coherency protocols describe rules for alternating processing of genetic sequence data stored in the shared memory based on a level of computational-intensity ascribed to a portion of the genetic sequence data stored in the memory.
3. The genomics analysis platform of claim 2 , wherein the rules for alternating processing of genetic sequence data stored in the shared memory include requiring use of the programmable logic circuit to process at least a first portion of the genetic sequence data stored in the shared memory, wherein the processing of the first portion of the genetic sequence data includes performing operations satisfying a threshold level of computational-intensity.
4. The genomics analysis platform of claim 3 , wherein the operations that satisfy the threshold level of computational intensity include the operations that account for 90% of the computing load.
5. The genomics analysis platform of claim 4 , wherein the operations that satisfy the threshold level of computational intensity include the operations that account for 98% of the computing load.
6. The genomics analysis platform of claim 2 , wherein the rules for alternating processing of genetic sequence data stored in the shared memory include requiring use of the processing unit to perform to process at least a second portion of the genetic sequence data stored in the shared memory, wherein the processing of the second portion of the genetic sequence data includes performing operations that fail to satisfy a threshold level of computational-intensity.
7. The genomics analysis platform of claim 3 , wherein the operations that fail to satisfy the threshold level of computational-intensity include the operations that account for 10% of the computing load.
8. The genomics analysis platform of claim 3 , wherein the operations that fail to satisfy the threshold level of computational-intensity include the operations that account for 2% of the computing load.
9. The genomic analysis platform of claim 2 , wherein the rules for alternating processing of genetic sequence data stored in the shared memory include:
identifying one or more computationally-intensive sub-routines of an operation,
using the programmable logic circuit to perform the identified one or more computationally-intensive sub-routines of the operations; and
using the processing unit to perform a balance of remaining sub-routines of the operation.
10. The genomic analysis platform of claim 2 , wherein the rules for alternating processing of genetic sequence data stored in the shared memory include:
for mapping and aligning operations, determining one or more discrete operations that are to be performed during the mapping and aligning operations;
triggering an acceleration call to the programmable logic circuit that instructs the programmable logic circuit to perform a subset of the discrete operations that satisfy a threshold level of computational-intensity; and
instructing the processing unit to perform a balance of the discrete operations.
11. The genomic analysis platform of claim 10 , wherein the one or more discrete operations include seed generation, mapping, seed chain formation, paired end rescue scans, gapless alignment, or gapped alignment.
12. The genomic analysis platform of claim 2 , wherein the rules for alternating processing of genetic sequence data stored in the shared memory include:
for variant calling operations, determining one or more discrete operations that are to be performed during the variant calling operations;
triggering an acceleration call to the programmable logic circuit that instructs the programmable logic circuit to perform a subset of the discrete operations that satisfy a threshold level of computational-intensity; and
instructing the processing unit to perform a balance of the discrete operations.
13. The genomic analysis platform of claim 12 ,
wherein the subset of the discrete operations that satisfy the threshold level of computational-intensity include using a pair hidden markov model to estimate a probability of observing a particular read given the truth of a candidate haplotype.
14. The genomic analysis platform of claim 1 , wherein the processing unit includes a central processing unit (CPU) or a graphical processing unit (GPU).
15. The genomic analysis platform of claim 1 , wherein the programmable logic circuit includes a field-programmable gate array (FPGA) or an Application Specific Integrated Circuit (ASIC).