RNA-seq is a robust method for quantifying gene expression levels based on short read data. As a standard service we offer two of the most popular workflows DESeq2 and Cufflinks. If you are unsure of which application best suites your research question be sure to check out the FAQ below or contact us for additional advice. We will happily help you choose the tool that best addresses your research question.

As a minimum we would recommend three replicates per treatment group. For an 'average' Eukaryotic genome the ENCODE standard recommendation is for a minimum of 20 million reads per replicate, however robust quantification of low expressed genes or alternative splice forms could require up to 70 million reads per replicate. For Prokaryote transcriptomes we recommend 5 million reads per replicate. Pair-end data is normally not required for prokaryote RNA-seq due to the absence of splicing. Please see the FAQ for additional experimental considerations.

Outputs your will receive with a typical project

  • Differential expression calls with statistical support
  • Functional annotations (if available)
  • Heatmaps of differentially expressed genes
  • Clustering and Principle component analysis of replicates and treatments
  • Aligned reads in the form of BAM files suitable for viewing using visulisation tools (for example IGV).

Frequently Asked Questions

As a minimum we would recommend three replicates per treatment group. Replicates are important because they help ascertain the level of within sample variation, which ultimately influences the number of significantly differential expressed genes that can be identified. Thus if you are working in a system where you expect large amounts of within sample variation, you may want to consider including additional replication.

ENCODE recommends at least 20 million reads per sample. If you are working with a non-model species you may want to consider additional reads to account for the possibility of low mapping rates.

At NgBS are huge fans of DESeq2 because we believe it takes a conservative approach in identifying differential expression and the authors are legends in the field of RNA-seq. However, we can also perform the analysis using the popular Cufflinks package. From a practical point of view, probably the biggest difference between the two packages is that Cufflinks is able to detect differential expression of novel genes and/or splice forms. In our experience DESeq2 tends to be more conservative and thus generally identifies fewer differentially expressed genes.

A 'standard project' (in terms of pricing) would involve a comparison between two treatment groups. The organism under investigation would also have a complete genome sequence available as well as a set of gene models (these are typically GFF or GTF files). If you are unsure if your research species has these resources just send us a quick inquiry and we should be able to let you know. If either of these files are missing additional work will be required to carry out the project and thus it will be slightly more expensive. Just get in contact with use and we should be able to give you a custom quote within two working days.

No! This is one of the advantages of RNA-seq over older microarray technologies. As a non-standard project we can take advantage of assemblers that include quantification pipelines (ie Trinity) or other custom approaches that we can discuss with you. Just send us a quick inquiry and we will typically get back to you within two working days.