RNA-Seq Analysis Generic Example: experiments with 192 samples or less

Costs circa 2021

Nanopore per flowcell ~$1,150.00 Nanopore library prep ~$125 BluePippin ~$200/sample for up to 10 cells

Costs circa 2017

Background

This example assumes genomic resources already exist for the organism under study. For example, the genome is assembled, annotated and available.

Consider the following experiment where an organism with 4 strains/lines/individuals is grown under 2 conditions (control and treated) with 3 time points and the experiment is biologically replicated 8 times for a total of 4 strains x 2 conditions x 3 time points x 8 replicates = 192 samples. The number of strains, conditions and time points can be changed by factors or combinations of factors of 24 (ie 2,3,4,6,8,12) depending on experimental design.

Currently, the max number of samples that can be indexed in a same lane for RNA-Seq is 24 and the max number of lanes in a flow cell is 8 (8x24=192). Therefore, each lane can act as a replicate of the entire experiment. This has the added advantage of avoiding lane effects and since all samples fit on a single flow cell, chip effects are also avoided.

An average estimate for the size of the genic space in a genome assumes a genome with ~30,000 genes with an average gene size of 1000 bases.

Ideal Experimental Design Elements for $136,000

  • Genome Assembled Yes
  • Cost for Sequencing Lanes: $22,720 (8 lanes * $2,840)
  • Cost for Library Prep: $31,680 (192 libraries * $165)
  • Total Sequencing Cost: $54,400
  • Cost for Bioinformatics: $81,600 (1.5 x sequencing cost)
  • Total Project Cost: $136,000
  • Sequencing Technology: Illumina HiSeq 3000
  • Assumed sequencing output: 300 million fragments/lane
  • Numbe r of lanes: 8 lanes
  • Length of read: 150bp
  • Number of Samples: 192 (4 strains x 2 conditions x 3 time points x 8 replicates)
  • Coverage Depth per sample: 12.5 million fragments per lane and on average ~416 fragments/gene
  • Number of Replicates: 8
  • Cost and technology as of: 2017

Ideal Experimental Design Elements for less than $20,000

  • Genome Assembled Yes
  • Cost for Sequencing Lanes: $2,840 (1 lanes * $2,840)
  • Cost for Library Prep: $3960 (24 libraries * $165)
  • Total Sequencing Cost: $6,800
  • Cost for Bioinformatics: $10,200 (1.5 x sequencing cost)
  • Total Project Cost: $17,000
  • Sequencing Technology: Illumina HiSeq 3000
  • Assumed sequencing output: 300 million fragments/lane
  • Number of lanes: 1 lanes
  • Length of read: 150bp
  • Number of Samples: 24 (1 strains x 3 conditions 8 replicates)
  • Coverage Depth per sample: 12.5 million fragments per lane and on average ~416 fragments/gene
  • Number of Replicates: 8
  • Cost and technology as of: 2017

1GB - 2GB Genome Assembly Generic Example

  • Genome Assembled No
  • Cost for Illumina Sequencing Lanes: $5,680 (2 lanes * $2,840)
  • Cost for Illumina Library Prep: $775 (2 libraries PE $105 + MP $670)
  • Cost of PacBio Sequencing: $17,280 (48 SMRT cells * $360)
  • Cost of PacBio Library Prep: $1050 (3 library preps)
  • Hi-C data: $15,000
  • Total Sequencing Cost: $39,785
  • Cost for Bioinformatics: $59,678 (1.5 x sequencing cost)
  • Total Project Cost: $99,463
  • Sequencing Technology: Illumina HiSeq 3000 + Pacbio RSII + HiC
  • Assumed sequencing output: 300 million fragments/lane + 500Mb/SMRT cell
  • Number of Illumina lanes: 2
  • Number of SMRT cells: 48
  • Length of Illumina read: 150bp
  • Length of Pacbio read: average around 12kb

less than 1 Gb Genome Assembly Generic Example (chromosomal level assembly)

  • Assumed Genome size 1Gb
  • Cost of PacBio Sequencing: $17,280 (48 SMRT cells * $360)
  • Cost of PacBio Library Prep: $1050 (3 library preps)
  • Hi-C data: $15,000
  • Total Sequencing Cost: $33,330
  • Cost for Bioinformatics: $49,995 (1.5 x sequencing cost)
  • Total Project Cost: $83,325
  • Sequencing Technology: Pacbio RSII
  • Assumed sequencing output: 500Mb/SMRT cell
  • Number of SMRT cells: 48
  • Length of Pacbio read: average around 12kb
  • Coverage depth: At least 24x coverage but more likely approaching 50x depending on size of genome and Pacbio Output

Table of contents