YaliFunTome

Yarrowia lipolytica Functional Screens of Tanscription Factor-ome Database

back home page

Data presentation

The TF co-OE strains were cultured under a combination of environmental variables which is schematically presented below. The data presentation scheme corresponds to the arrangement of conditions shown in Fig.1:

Experimental setup figure

Fig.1. Schematic representation of the variables combinations - corresponding to the data presentation scheme.

Each strain overexpressing an individual TF (TF-OE strain) was tested under an array of conditions, and so was the reference strain. Raw data for:

  • optical density at 600 nm wavelength as a proxy of growth,
  • fluorescence intensity from a reporter protein as a proxy of r-Prot synthesis,
were processed according to the Response Surface Methodology to acquire models describing a behavior of a given read parameter within the range of investigated variables. An exemplary set of Response Surface Models for a single parameter investigated for a single strain is graphically presented in Fig.2. The set of eight graphs is arranged according to design of experiment presented in Fig.1.

Graphs representing Response Surface Models

Fig.2. Set of graphs representing Response Surface Models of a single read parameter (growth in this example) under an array of variables combinations - arrangement of the graphs corresponds to Fig.1. Schematic representation of the combinations of the variables.

The final results are presented as heat-maps (Fig.3) of fold change (FC) of a given measure response in the TF-OE strain over the control strain.

Example of heat map

Fig.3. Heat map indicating Fold Change in a given parameter (growth / r-Prot / normalized r-Prot) for a specific TF-OE strain over the control strain under a specific combination of the variables; arrangement corresponds to the experimental setup scheme.

The adopted data processing strategy enabled the evaluation of each specific variable's contribution to the response's variability (growth, amount of r-Prot) elicited by a given TF-OE strain. This evaluation is presented in Factor's Contribution tables indicating the dominant level of a variable (color-coded and as +1 or -1, according to the adopted coding system) and a percentage contribution of the variable to the response. Exemplary data showing the ranking of variables' contribution to the analyzed response is shown in Fig.4. with a reminder of the variables' coding system.

Factor's Contribution ranking
  • ■ and contribution % value > 0: dominant level +1, variable's contribution higher than in the control strain
  • ■ and contribution % value < 0: dominant level +1, variable's contribution lower than in the control strain
  • ■ and contribution % value > 0: dominant level -1, variable's contribution higher than in the control strain
  • ■ and contribution % value < 0: dominant level -1, variable's contribution lower than in the control strain
  • □ non-significant contribution of the variable's
  • ↕ impact inversion: the preferred level of the variable is opposite in the control strain
  • ↔ impact gain: impact is non-significant in the control strain, and gained significance in the TF-OE strain

Fig.4. Graphical presentation of the variables' coding system and Exemplary Factor's Contribution ranking with a legend. The coding system corresponds to Tab.1 in experimental setup CH- casamino acid hydrolysate, AS - ammonium sulfate, OA- oxygen availability.

The rankings are prepared based on Response Surface Models (Fig.2 in data presentation ). Each ranking represents variable's contribution to a response of a given read parameter (growth / r-Prot / normalized r-Prot) elicited by a specific TF-OE strain.