Can someone read this paper?

[research555]

I've read it three times, highlighted many different sentences and I still dont understand a single word of what is going on. Can one of you assign yourselves to have a crack at it?

https://www.nature.com/articles/nature12148

[JulietteB-cri]

From my first lecture - I will complete it after my future relecture.

• They performed analog computations with only a few biological parts of a living cells
• You can build a ratiometer with only 2 or 3 TF
• You can use digital circuits to integrate sensors and respond to environmental conditions and analogue circuits to do arithmetic functions with a small number of regulators
• They did logarithmically linear computation in biochemical systems (energy efficient), can be applied to brain circuits.
• You can implement PF in circuits such as cortical circuits - that follows the Weber law
• PF are essential to increase the log-linear dynamic range of operations

[research555]

still speaking french to me lmao. i guess im just an idiot

[Zoepin]

I just read the abstract and faced my first big question: what is Weber's law? Weber’s law is a historically important psychological law quantifying the perception of change in a given stimulus. The law states that the change in a stimulus that will be just noticeable is a constant ratio of the original stimulus. It has been shown not to hold for extremes of stimulation. https://www.britannica.com/science/Webers-law

Then what is analog computation? An analog computer or analogue computer is a type of computer that uses the continuously changeable aspects of physical phenomena such as electrical, mechanical, or hydraulic quantities to model the problem being solved. In contrast, digital computers represent varying quantities symbolically and by discrete values of both time and amplitude.

Figure 1 They created a biological loop with an original input of x (arabinose) and an output concentration of ln(x+1) araC with a wide dynamic range. Instead of having a threshold arabinose concentration at which the output starts being produced at a single concentration, they implemented various feedbacks so that a small variation in arabinose as an input will change the concentration of the output and therefore obtain an analog function. To do that, they insured the promoters would never be saturated and would be able to detect small changes of their binding transcription factors.

• They then obtained similar results but changed the araC gene to LuxR for quorum sensing.

• They then tried to widen the log-linear dynamic range of operation by means of graded positive feedback. (Basically they tried to obtain different versions of that first ln(1+x).

Figure 2 The behaviour of the created (PFS) circuit motif can be dynamically tuned by changing the relative copy numbers of the positive-feedback and shunt plasmids.

So they changed the copy numbers and included variable copy number (VCN) plasmids and obtained different curves all assimilated to ln(1+x) but with different dynamic ranges.

• They tried to obtain higher order functions (other functions then ln)

Figure 3 a - If you combine 2 positive-logarithm circuits (figure 1) with different inputs (arabinose and AHL) but the same output mCherry, then the output production follows a negative slope ln.

c - If you combine a positive-logarithm circuit (figure 1) with a negative-logarithm circuit (figure 3) then you get a ratiometer. The concentration of the output is the log of the ratio of the positive loop input by the negative loop output.

e - By creating a strong negative loop, you can obtain power laws.

[JulietteB-cri]

Small additions to my first reading that I think are relevant

Why analog?

• because compared to digital computation, this computation mode is more efficient in terms of speed, energy consumption....

How can we reduce the variation of cross-over precision due to computation?

• Using feedbacks loops or signal averaging to increase precision for example