Portable nanopore analytics: are we there yet?
basic_science · Level V
Where this comes from
- Record sourced from PubMed, PMID 32277811.
- Also identified by DOI 10.1093/bioinformatics/btaa237 and PMC identifier 7828464.
- No licence information is recorded for this record.
- Because redistribution is not established, this page shows the abstract only. Follow the links below for the full text.
Abstract
Oxford Nanopore technologies (ONT) add miniaturization and real time to high-throughput sequencing. All available software for ONT data analytics run on cloud/clusters or personal computers. Instead, a linchpin to true portability is software that works on mobile devices of internet connections. Smartphones' and tablets' chipset/memory/operating systems differ from desktop computers, but software can be recompiled. We sought to understand how portable current ONT analysis methods are. Several tools, from base-calling to genome assembly, were ported and benchmarked on an Android smartphone. Out of 23 programs, 11 succeeded. Recompilation failures included lack of standard headers and unsupported instruction sets. Only DSK, BCALM2 and Kraken were able to process files up to 16 GB, with linearly scaling CPU-times. However, peak CPU temperatures were high. In conclusion, the portability scenario is not favorable. Given the fast market growth, attention of developers to ARM chipsets and Android/iOS is warranted, as well as initiatives to implement mobile-specific libraries. The source code is freely available at: https://github.com/marco-oliva/portable-nanopore-analytics.
Medical subject headings
- Nanopores