# supplementary information - media. supplementary information 2 | research supplemental figure2...

Post on 10-Jun-2018

213 views

Embed Size (px)

TRANSCRIPT

w w w. n a t u r e . c o m / n a t u r e | 1

SuPPLementarY InFormatIondoi:10.1038/nature09679

Supplemental Figure 1Regot et al.

Figure S1. Illustrative example of the method for designing distributed biological computation. a. General pattern of connections between cells for multicellular circuits.b. Truth table defining the behavior of a multiplexer circuit. c. Full Boolean function that express the relationship between the logic inputs x1, x2, x3 and the logic output O according to the truth table. d. Simplified version of the Boolean function f describing the same truth table. The analytical expression of the Boolean function determines that the circuit can be implemented using two different sets of engineered cells 1 and 2. e. Implementation of the circuit with direct mapping between the terms of the Boolean function and the engineered cells. Here 1={C11, C12} and 2={C21, C22}. The tables determine the logic implemented in each cell. Here rhombus represents the external inputs x1, x2, and x3, red and blue circles represent the internal wiring implemented by the production of a diffusible molecule, and squares are the final output. f. A mixed implementation, where the set 1 is a direct mapping of the terms from the Boolean function whereas in 2 all the terms have been condensed in a single cell. g. Circuit implementation where both sets 1 and 2 are implemented by a single cell in each case. Figure S2. Number of possible Boolean functions versus the number of different cells required for their implementation. Each graph represents the number of (nonnull) functions that can be implemented with a defined number of engineered cells that receive 2-inputs (a) and 3-inputs (b). Figure S3. Mathematical analysis of possible 3-input 1-output Boolean functions versus the number of different wires required for their implementation upon different approaches. a. Multicellular approach, b. standard approach based on NAND logic, c. standard approach based on NOR logic

Figure S4. Complete description of a engineered yeast cell library. a. Graphical respresentation and basic genetic information of the two main types of cells used. Schematic representation of cells that express a wiring molecule (Left) and sensor cells (Right). Schematic legend of logic functions. b. Schematic representation of each cell of the library indicating the logic function performed by each cell and their corresponding genotype (all cells are W303 derivatives). Note that some cells can implement two different functions depending on the experimental conditions and the input used (in brackets). Figure S5. Transfer function analyses of the engineered cells of the library. Indicated cells were grown in YPD to mid exponential phase. Cells that respond to a single input were mixed with the appropriate reporter cell (carrying S. cerevisiae or C.albicans STE2) and treated with variable amounts of input (cells treated with synthetic alpha factor were included as a reference). Output producing cells were mixed with appropriate synthetic alpha factor (from S. cerevisiae or C.albicans) and the indicated input, both at different concentrations. Samples were incubated for 4h at 30C and analyzed by flow cytometry as described in methods. Data is expressed as the percentage of GFP positive cells and represent the mean of three independent experiments. Data are presented as a regular graph for 1 input analyses and as a contour plot for 2 input analyses.

SUPPLEMENTARY INFORMATION

2 | w w w. n a t u r e . c o m / n a t u r e

RESEARCH

Supplemental Figure2Regot et al.

a

bFigure S1. Illustrative example of the method for designing distributed biological computation. a. General pattern of connections between cells for multicellular circuits.b. Truth table defining the behavior of a multiplexer circuit. c. Full Boolean function that express the relationship between the logic inputs x1, x2, x3 and the logic output O according to the truth table. d. Simplified version of the Boolean function f describing the same truth table. The analytical expression of the Boolean function determines that the circuit can be implemented using two different sets of engineered cells 1 and 2. e. Implementation of the circuit with direct mapping between the terms of the Boolean function and the engineered cells. Here 1={C11, C12} and 2={C21, C22}. The tables determine the logic implemented in each cell. Here rhombus represents the external inputs x1, x2, and x3, red and blue circles represent the internal wiring implemented by the production of a diffusible molecule, and squares are the final output. f. A mixed implementation, where the set 1 is a direct mapping of the terms from the Boolean function whereas in 2 all the terms have been condensed in a single cell. g. Circuit implementation where both sets 1 and 2 are implemented by a single cell in each case. Figure S2. Number of possible Boolean functions versus the number of different cells required for their implementation. Each graph represents the number of (nonnull) functions that can be implemented with a defined number of engineered cells that receive 2-inputs (a) and 3-inputs (b). Figure S3. Mathematical analysis of possible 3-input 1-output Boolean functions versus the number of different wires required for their implementation upon different approaches. a. Multicellular approach, b. standard approach based on NAND logic, c. standard approach based on NOR logic

Figure S4. Complete description of a engineered yeast cell library. a. Graphical respresentation and basic genetic information of the two main types of cells used. Schematic representation of cells that express a wiring molecule (Left) and sensor cells (Right). Schematic legend of logic functions. b. Schematic representation of each cell of the library indicating the logic function performed by each cell and their corresponding genotype (all cells are W303 derivatives). Note that some cells can implement two different functions depending on the experimental conditions and the input used (in brackets). Figure S5. Transfer function analyses of the engineered cells of the library. Indicated cells were grown in YPD to mid exponential phase. Cells that respond to a single input were mixed with the appropriate reporter cell (carrying S. cerevisiae or C.albicans STE2) and treated with variable amounts of input (cells treated with synthetic alpha factor were included as a reference). Output producing cells were mixed with appropriate synthetic alpha factor (from S. cerevisiae or C.albicans) and the indicated input, both at different concentrations. Samples were incubated for 4h at 30C and analyzed by flow cytometry as described in methods. Data is expressed as the percentage of GFP positive cells and represent the mean of three independent experiments. Data are presented as a regular graph for 1 input analyses and as a contour plot for 2 input analyses.

w w w. n a t u r e . c o m / n a t u r e | 3

SUPPLEMENTARY INFORMATION RESEARCHSupplemental Figure 3

Regot et al.

a

b

c

Figure S1. Illustrative example of the method for designing distributed biological computation. a. General pattern of connections between cells for multicellular circuits.b. Truth table defining the behavior of a multiplexer circuit. c. Full Boolean function that express the relationship between the logic inputs x1, x2, x3 and the logic output O according to the truth table. d. Simplified version of the Boolean function f describing the same truth table. The analytical expression of the Boolean function determines that the circuit can be implemented using two different sets of engineered cells 1 and 2. e. Implementation of the circuit with direct mapping between the terms of the Boolean function and the engineered cells. Here 1={C11, C12} and 2={C21, C22}. The tables determine the logic implemented in each cell. Here rhombus represents the external inputs x1, x2, and x3, red and blue circles represent the internal wiring implemented by the production of a diffusible molecule, and squares are the final output. f. A mixed implementation, where the set 1 is a direct mapping of the terms from the Boolean function whereas in 2 all the terms have been condensed in a single cell. g. Circuit implementation where both sets 1 and 2 are implemented by a single cell in each case. Figure S2. Number of possible Boolean functions versus the number of different cells required for their implementation. Each graph represents the number of (nonnull) functions that can be implemented with a defined number of engineered cells that receive 2-inputs (a) and 3-inputs (b). Figure S3. Mathematical analysis of possible 3-input 1-output Boolean functions versus the number of different wires required for their implementation upon different approaches. a. Multicellular approach, b. standard approach based on NAND logic, c. standard approach based on NOR logic

Figure S4. Complete description of a engineered yeast cell library. a. Graphical respresentation and basic genetic information of the two main types of cells used. Schematic representation of cells that express a wiring molecule (Left) and sensor cells (Right). Schematic legend of logic functions. b. Schematic representation of each cell of the library indicating the logic function performed by each cell and their corresponding genotype (all cells are W303 derivatives). Note that some cells can implement two different functions depending on the experimental conditions and the input used (in brackets). Figure S5. Transfer function analyses of the engineered cells of the library. Indicated cells were grown in YPD to mid exponential phase. Cells that respond to a single input were mixed with the appropriate reporter cell (carrying S. cerevisiae or C.albicans STE2) and treated with variable amounts of input (cells treated with synthetic