About Standards and Replicon Genetics …

…objectives

Our aim is to create synthetic sequence standards that are useful for:

• The creation of test data, such as ‘spike-in controls’, by Bioinformaticians managing variant-calling bioinformatics pipelines
• Diagnostic laboratories running such pipeline
• Diagnostic developers validating their NGS bioinformatics pipelines
• Situations where real tissue samples are unavailable
• Students of bioinformatics creating bioinformatics pipelines and needing test datasets to work with.

We want you to have confidence that your bioinformatics pipelines are giving accurate insights from real patient data and to help you to support accurate patient diagnosis.

References about standards:

Hardwick, S., Deveson, I. & Mercer, T. Reference standards for next-generation sequencing. Nat Rev Genet 18, 473–484 (2017). https://doi.org/10.1038/nrg.2017.44

Blackburn, J., Wong, T., Madala, B.S. et al. Use of synthetic DNA spike-in controls (sequins) for human genome sequencing. Nat Protoc 14, 2119–2151 (2019). https://doi.org/10.1038/s41596-019-0175-1

O’Sullivan, D.M., Doyle, R.M., Temisak, S. et al. An inter-laboratory study to investigate the impact of the bioinformatics component on microbiome analysis using mock communities. Sci Rep 11, 10590 (2021). https://doi.org/10.1038/s41598-021-89881-2

.…personalized standards for you

Synthetic Reads Generator (SyRGen) can be customised to generate a set of standards suitable for your analysis pipeline:

The provided haplotype data here can be used to create any spike-in control; using the source code, anyone can create a custom set of haplotypes to represent different germline and somatic genotypes arising from normal and tumour cell lines (such as might result from a tissue sample). You can then use the browser-based app or TKinter GUI to adjust the following parameters:

• Depth of coverage
• Read-length
• Relative proportion of haplotypes
• Range of data quality
• Genomic, mRNA (“exome”) and CDS data

Other options support:

• Paired-end sequencing; dual-strand or single-strand sequencing
• Selection of variant-only reads
• Genomic-position annotation
• CIGAR annotation
• Annotation of haplotype source
• Randomly generated reads vs all-possible reads (random-only for paired-end).

See the Help Page for more detail

Services

This work is in the Public Domain.

We are no longer looking to provide any services, beyond enabling collaborative development in the Public Domain.

For suggestions and further development, please contact us at

syrgenreads@gmail.com

…origins

Replicon Genetics was set up on 21 August 2018 by two former-AZ scientists who saw the need for simulated NGS data standards in precision medicine to validate bioinformatics pipelines. Having worked with hospital diagnostic laboratories they could see that sequence-data standards were being generated from wet-chemistry standards but often did not contain the mutations of interest, with variables such as data quality, coverage and other parameters uncontrolled.

Hearing comments like: “We don’t have a problem with our bioinformatics pipelines; it’s all those other laboratories with dodgy, poorly-validated pipelines that have the problem.” was another incentive.

But they wondered “How can you be sure that your bioinformatics pipeline isn’t misdiagnosing hundreds of patients each week? Quality isn’t a problem…. until it’s a BIG problem”

After a conversation at Dunham Massey National Trust coffee shop, a former-AZ bioinformatician agreed to address the task on a “part-time basis”.

To avoid hosting large sequence-read files on a website, with the inevitable problem of slow download speeds, an obvious solution was a browser-based generator of reads that we named “Synthetic Reads Generator“; succinctly SyRGen. This is built on top of Python code called RG_exploder.