Sumary of Library preparation

mRNA Seq

This technique converts poly-adenylated RNA from total RNA populations into libraries of template molecules of known strand origin, which are then suitable for sequencing on Illumina platforms.

Protocol used by Leeds NGS Facility

TruSeq Stranded mRNA Sample Preparation Kit (Illumina).

Overview:

HiSeq2500 run parameters

Typically 4-6 samples are run per lane of HiSeq2500 paired end 100 bp run, resulting in approx. 200M reads passing filter.
Samples numbers can be changed depending on the depth of coverage required.
Fresh samples
Ideally 30M mapping reads are required for analysis and based on 80 % of raw reads mapping, means 40 M raw reads are required per sample.
samples
The same number of mapping reads are required for analysis (30M), but only approx. 70 % of reads map, thus 45M raw reads are required per sample.

Total RNA Seq

This technique converts total RNA into libraries of template molecules of known strand origin, which are then suitable for sequencing on Illumina platforms.

Protocol used by Leeds NGS Facility

TruSeq Stranded Total RNA Preparation Kit (Illumina).

Overview:

HiSeq2500 run parameters

Typically 3-4 samples are run per lane of a HiSeq2500 paired end 100 bp run resulting in approx. 200M reads passing filter (conservative estimate).
Samples numbers can be changed depending on the depth of coverage required.
Fresh samples
Ideally 50M mapping reads are required for analysis and based on 80 % of raw reads mapping, means 65 M raw reads are required per sample.
FFPE samples
The same number of mapping reads are required for analysis (50M), but only approx. 70 % of reads map, thus 70M raw reads are required per sample.

miRNA Seq

This technique captures and enriches miRNA from total RNA populations into libraries of template molecules, which are then suitable for sequencing on Illumina platforms.

Protocol used by Leeds NGS Facility

NEBNext Small RNA Library Prep for Illumina.

Overview:

HiSeq2500 run parameters

Up to approx. 40 samples can be run per lane of a HiSeq2500 single read 50 bp run, resulting in approx. 200M reads passing filter (if optimal cluster densities are achieved).
Ideally 250,000 mapping reads per sample are sufficient for analysis, but given the huge diversity in the literature regarding the number of reads people see mapping, we recommend 5M raw reads per sample to gain the 250,000 mapping reads. In our limited experience of miRNA Seq, we have seen even lower mapping statistics for FFPE samples compared to fresh, so for these samples, we would recommend pooling fewer samples to achieve a higher number of raw reads per sample.