Then again, I do believe that you can culture simple cells for a lower cost. I estimated the cost of producing protein-rich single cells from methane at $1-2/dry kg. https://osf.io/94mkg/
However, those numbers are for a bacteria that feeds on gas. The yeast analogy is much closer to mammalian cultures.
Oh very cool! I know you/ALLFED is focused on bacterial SCPs from the perspective of feeding earth after disasters, but do you have a quick sense of why we can’t have it during normal times? Like what are the relevant timelines/bottlenecks against using bacterial SCPs as a way to partially replace animal agriculture?
Also more broadly did you do any BOTECs to look at the cost of yeast SCPs in comparison to bacterial SCPs? Just generally curious if we actually expect bacterial SCPs to have cost savings over yeast (waving away scientific problems)!
Pretty much the only thing currently standing between us and bacterial SCP-based food (such as from methane or CO2/H2) is the lack of approval for use as a human food. Most or all of these companies have shown interest in the human food market, and a few of them are publicly pursuing it, such as Solar Foods. I expect they will be available in the next few years.
As Humbird mentions in the TEA and other sources confirm, the production cost of baker’s yeast is well known (~$1.80/kg dry), so no need to run any numbers for that. I’m fairly confident SCP from methane will be lower cost at scale as I mentioned above, but not SCP from CO2. Methane SCPs will almost certainly be quite a bit cheaper than yeast in terms of cost per unit of protein.
Fun fact: these protein-rich SCPs will be so much cheaper than cultured meats, that SCP companies have been promoting their use as a raw material for cultured meat production.
As Humbird mentions in the TEA and other sources confirm, the production cost of baker’s yeast is well known (~$1.80/kg dry), so no need to run any numbers for that
I think there’s two missing steps here.
1) yeast SCP for vegan meat replacement(Quorn) probably has importantly different processes and structure to baker’s yeast. So I’m interested in whether it’s realistic to expect ~$2/kg for that as well.
2) Humbird’s TEA for yeast has both internal validity and empirical validation, and also he made a yeast TEA as a validation step to demonstrate that his model can retrodict an existing system. The secondary reason I asked for you to share the same methods that you’ve used for bacterial SCPs to get to yeast SCP numbers is so we can help make sure you aren’t skipping calculation steps.
(But it’s probably not worth you doing additional work here if you haven’t already done so, just for my curiosity)
I think I see what you’re getting at, let me add a couple of things:
Quorn’s mycoprotein is produced from a different microorganism (Fusarium Venenatum), with different growth rates and processing steps than baker’s yeast (Saccharomyces cerevisiae), so you are correct. It is more expensive than yeast, and also compared to the gas-based SCPs I mentioned.
Based on a conversation I had with an ex-Quorn scientist, the wholesale selling price of Quorn products is ~$3/kg wet (which makes sense given the intensive postprocessing and other additional steps). I’m uncertain as to how it would differ from yeast when comparing them in a similar state (e.g. drying as pretty much the sole post-processing step, which is the case of the values I gave for yeast and methane-based SCP), but I imagine that even then Quorn would be at least twice the production cost of yeast/methane SCP.
A little clarification about my research: I did not run a TEA as such to calculate capital and operational expenditures of the gas-based SCP production systems. Rather, I used published values from TEAs run by SCP companies themselves as an input to my model. I used these for my goal of estimating the retail price in a catastrope and the ramp-up speed of fast construction of SCP factories in case of catastrophe.
I did not try to perform validation of the published values as I did not have access to the calculations run by those companies. Instead, I made sure that my results are robust to relatively large changes in the TEA results published by the SCP companies via sensitivity analysis. I’m happy to talk more about the methodology if you’re interested.
Then again, I do believe that you can culture simple cells for a lower cost. I estimated the cost of producing protein-rich single cells from methane at $1-2/dry kg. https://osf.io/94mkg/
However, those numbers are for a bacteria that feeds on gas. The yeast analogy is much closer to mammalian cultures.
Oh very cool! I know you/ALLFED is focused on bacterial SCPs from the perspective of feeding earth after disasters, but do you have a quick sense of why we can’t have it during normal times? Like what are the relevant timelines/bottlenecks against using bacterial SCPs as a way to partially replace animal agriculture?
Also more broadly did you do any BOTECs to look at the cost of yeast SCPs in comparison to bacterial SCPs? Just generally curious if we actually expect bacterial SCPs to have cost savings over yeast (waving away scientific problems)!
Pretty much the only thing currently standing between us and bacterial SCP-based food (such as from methane or CO2/H2) is the lack of approval for use as a human food. Most or all of these companies have shown interest in the human food market, and a few of them are publicly pursuing it, such as Solar Foods. I expect they will be available in the next few years.
As Humbird mentions in the TEA and other sources confirm, the production cost of baker’s yeast is well known (~$1.80/kg dry), so no need to run any numbers for that. I’m fairly confident SCP from methane will be lower cost at scale as I mentioned above, but not SCP from CO2. Methane SCPs will almost certainly be quite a bit cheaper than yeast in terms of cost per unit of protein.
Fun fact: these protein-rich SCPs will be so much cheaper than cultured meats, that SCP companies have been promoting their use as a raw material for cultured meat production.
I think there’s two missing steps here.
1) yeast SCP for vegan meat replacement(Quorn) probably has importantly different processes and structure to baker’s yeast. So I’m interested in whether it’s realistic to expect ~$2/kg for that as well.
2) Humbird’s TEA for yeast has both internal validity and empirical validation, and also he made a yeast TEA as a validation step to demonstrate that his model can retrodict an existing system. The secondary reason I asked for you to share the same methods that you’ve used for bacterial SCPs to get to yeast SCP numbers is so we can help make sure you aren’t skipping calculation steps.
(But it’s probably not worth you doing additional work here if you haven’t already done so, just for my curiosity)
I think I see what you’re getting at, let me add a couple of things:
Quorn’s mycoprotein is produced from a different microorganism (Fusarium Venenatum), with different growth rates and processing steps than baker’s yeast (Saccharomyces cerevisiae), so you are correct. It is more expensive than yeast, and also compared to the gas-based SCPs I mentioned.
Based on a conversation I had with an ex-Quorn scientist, the wholesale selling price of Quorn products is ~$3/kg wet (which makes sense given the intensive postprocessing and other additional steps). I’m uncertain as to how it would differ from yeast when comparing them in a similar state (e.g. drying as pretty much the sole post-processing step, which is the case of the values I gave for yeast and methane-based SCP), but I imagine that even then Quorn would be at least twice the production cost of yeast/methane SCP.
A little clarification about my research: I did not run a TEA as such to calculate capital and operational expenditures of the gas-based SCP production systems. Rather, I used published values from TEAs run by SCP companies themselves as an input to my model. I used these for my goal of estimating the retail price in a catastrope and the ramp-up speed of fast construction of SCP factories in case of catastrophe.
I did not try to perform validation of the published values as I did not have access to the calculations run by those companies. Instead, I made sure that my results are robust to relatively large changes in the TEA results published by the SCP companies via sensitivity analysis. I’m happy to talk more about the methodology if you’re interested.