Many factors come into play when considering the purchase of a sequencing machine: machine cost; reagent cost; and sometimes compute cost. Many times the machine’s utilization rate is not factored into the decision. This failure to objectively determine the true cost per run often leads to a gross misallocation of resources.

Average Total Cost is distinct from price and the variable cost. ATC is the cost per unit of output, and its natural presence is woven into the fabric of modern economic theory. It can be found explicitly by dividing the total cost to produce by the quantity of output. ATC is a standard measurement of scale economies used by economists and managers to view production efficiencies and profitability as a function of quantity. Average total cost is minimized when workflow is constant enough to push the machine to its temporal and physical limits (ie. each run has so much capacity, and each year only so many runs can be completed). In other words, one machine can only do so much.

The costs associated with owning & operating a sequencing platform are numerous, and are often overlooked by decision-makers. These costs include:

• Initial purchase
• Installment and shipping fees
• Service agreement
• Compute server and storage
• Labor expenses to operate machine and analyze the data
• Reagent cost per run, dependent upon run type

There are also less tangible costs such as opportunity costs, constraints associated with running a single platform, machine downtime and machine failure. The most elusive cost is somewhat intangible: organizational resources are diverted away from value-adding core competencies. This post will keep to the more tangible and direct fiscal costs bulleted above.

One of the most informative ways to consider this production phenomenon is to look at it in the context of two common scenarios.

Scenario 1: A research lab wants to buy a “small, cheap” platform for next year’s sequencing. The platform costs $125,000, and then $900 per run for library prep and sequencing reagents after that; not bad, right? This lab has 24 candidates flowing through their group that year and plans to sequence each candidate. This is great, because 24 runs are well within the capability of one of these “small, cheap” platforms and they will not have to invest in additional personnel to run the machine or handle the analysis because it is a relatively small number of runs. However, the fixed costs are only allocated over 24 runs… what does this do to the ATC? On a per run basis (ATC), even while excluding setup, delivery, and service agreement expenses, this lab spent about $6,000 per run; orders of magnitude more expensive than current outsourcing options. This is highlighted in Figure A.

[$125,000 + $900(24 runs)] / 24 runs = $6,108 per run

Figure A.

NOTE: Service agreement, installment, shipping, and labor costs are excluded from consideration in Figure A.

Scenario 2: Another common scenario is an industry group that is ramping up their discovery pipeline and they plan on having 120 candidates to work through next year. Given the number of samples they are dealing with, it surely must be cheaper to sequence these samples in house rather than outsource, right? This is 10 samples a month for a year straight. In this scenario, highlighted in Figure B, expenses associated with a service agreement, initial installation, and both a lab and bioinformatician are considered in the total cost.

Figure B.

We have found that the institutions looking at this most critically choose to outsource due to many negative aspects associated with seting up capacity in-house. Factors associated with an in-house approach are; the relative high cost of entry, the fact that the industry is innovating and changing on average every 6 months or less, the instrumentation requires specialized personnel, the lag in obtaining useful data, and the high cost due to low sample number and not achieving economies of scale. What the ATC tables and above scenarios focus on are this last component (economies of scale). Cofactor achieves economies of scale due to the hundreds of clients we work with and our clients benefit from this cost savings as well as the benefit of the experience our team has due to the thousands of sequenced samples under their belt.

 

While there may be reasons for groups to house and run their own sequencing instrument as opposed to outsourcing their workflow… lower cost is rarely the reason to take the plunge.