New web site is available.
Phenodyn movie is available.
Arabidopsis thaliana phenotyping database
Maize field trials database
Analysis and modelling of the genotype x environment interaction
Expansive growth of leaves or of reproductive organs such as silks are affected by water deficit before any reduction in photosynthesis or root growth. The objective of the Phenodyn platform is to perform a genetic analysis of growth and gas exchanges which vary rapidly with environmental conditions (see movie). In particular, we aim to disentangle the genetic basis of the differences in growth rate and of its responses to temperature, evaporative demand and soil water deficit. Time courses of elongation rate, transpiration and environmental conditions are dissected into more heritable traits (maximum rate and sensitivity) which are stable characteristics of genotypes, amenable to genetic analysis. Phenodyn has been used for detection of QTLs of sensitivity of leaf elongation rate to soil water deficit and evaporative demand1,2, for physiological analyses using transgenic or introgression lines5, and for analysing the genetic variability of growth and of its sensitivity to environmental conditions in panels of lines with different geographical and genetic origins. Phenodyn has been tested for maize leaves 1,2,3,5 and rice leaves 6 and for maize silks 4.
The Phenodyn platform (current throughput : 480 plants) imposes drought scenarios under fluctuating atmospheric conditions in the greenhouse or the growth chamber, and follows transpiration and expansive growth with a definition of 5-15 min over periods of 1-2 weeks. This allows identification of the genetic variability of response of growth and transpiration to environmental cues, and in a quantification of the sensitivity to abiotic stresses. The experimental set-up in the greenhouse consists in 140 balances which measure changes in soil water status and transpiration, 420 displacement transducers which continuously measure leaf elongation rate, and a set of climatic sensors. A companion set-up is placed in a growth chamber, with the same sensors for measurement of 60 plants simultaneously. Leaf or silk elongation rates are measured with rotational displacement transducers. They are transmitted to the sensor via a pulley which carries a thread attached to the leaf or silk tip and to a 20 g counterweight. Soil water content is estimated by continuously weighing columns. Differences in weight are attributed to changes in soil water content, after correction for the increase in mean plant biomass as a function of phenological stage. Soil water content is transformed into soil water potential via a water release curve corresponding to the potting compost. Air temperature, relative humidity and light are measured at plant level, with two series of sensors per block. The temperature of the meristematic zone of studied organs is measured with fine copper-constantan thermocouples. All data are averaged and stored every 5 to 15 min in the database. It was checked (i) that the procedure for estimating soil water content generates errors smaller than 3 g , i.e. an error in soil water content of about 6 10-4g, (ii) that a counterweight of 20 g does not cause changes in elongation rate of leaf or silks of maize. Conversely, it affects leaf growth of rice on the first day after emergence.
A/vesala sensor B-C/rotating displacement transducers D/precision weigthing balances E/thermocouple
The phenotyping platform is associated to an information system for real time monitoring of experiments, post-analysis of large datasets (around 700.000 datapoints are generated in each experiment) and identification of genotypic parameters.
Two computers automatically send data to a database through wifi access every 5 to 15 minutes. Manual measurements on plants (ie. phenology) are also entered in the database using CSV files. Data are stored in an MySQL 5.0 database hosted on a Linux server. Database interface is based on PHP programs or HTML pages and some R scripts to generate elaborated graphics. We also use Ajax for a better user interaction.
PHENODYN data acquisition workflow.
The system is based on a client-server architecture. Time-courses are visualised using PHP (jpgraph library) or R scripts.
(A) plant growth rate, (B) micro-meteorological conditions sensed by the plants (meristem temperature), (C) soil moisture content of the pot in which plant grows up, (D) leaf or silk elongation rates.
The data base stores information on (i) the experimental design, (ii) the genetic material and data collected over the experiments (iii) meteorological variables, (iv) weight of pot and estimated soil water content (v) elongation rates or daily information such as (vi) development stage of plants.
Each experiment is identified by a code (table Manip). Each experiment can contain pots (table Pot). Each pot contains plants (table Plante). Each plant has a genotype (table Genotype). Three type of variables are measured: Meteorological variables (table mesureMeteo), elongation rates (table MesureCroissance), phenological stages (leaf number) estimated by visual inspection (table Phenology).
PHENODYN database physical model.
Time courses in the greenhouse (left) or growth chamber (right) of leaf elongation rate per unit thermal time (LER). Lower panel, mean and interval of confidence of 40-60 time courses). Upper panel : evaporative demand (VPD, green or blue) and light (red), transpiration rate (black) every 15 min.
Silk growth is affected by evaporative demand (black arrows) and soil water deficit (red arrows). Inbred lines with short ASI in dry fields maintained silk growth under water deficit.
Joint analysis of elongation rate and transpiration under moderate water deficit showed that (i) during the night, ABA+ lines had a lower elongation rate than WT, (ABA- a higher LER). (ii) During the day, lower transpiration rate in ABA+ lines resulted in higher water status and a higher elongation rate (opposite for ABA- lines) (iii) all transitions were quickest in ABA+ lines, suggesting higher hydraulic conductance.
A genetic analysis is carried out on the model parameters (maximum elongation rate, responses to evaporative demand and to soil water deficit. This provides QTLs of parameters with no genotype x environment interaction. This also allows calculating the parameters values corresponding to any genotype of the population, from its allelic composition. The behaviour of these "virtual genotypes" can then be simulated in any climatic scenario.
Database is available for internal use only. Please contact Claude Welcker for further information.
- Scientific managers : Tardieu François, Welcker Claude
- Technical managers : Berthezene Stéphane, Brichet Nicolas, Hamard Philippe, Suard Benoît
- Informatics: Vincent Negre,Tireau Anne, Pascal Neveu
© 2007-2010 INRA LEPSE-MISTEA | credits | Legal notice | | Home page