Research engineer M/F in biostatistics applied to ecotoxicology, VILLEURBANNE, France



General information

Reference : UMR5558-SANCHA-002
Date of publication : Friday, March 22, 2019
Type of Contract : FTC Technical / Administrative
Contract Period : 12 months
Expected date of employment : 3 June 2019
Proportion of work : Full time
Remuneration : Between 2423€ and 2612€ of monthly gross pay according to experience
Desired level of education : Higher than 5-year university degree
Experience required : 1 to 4 years


The main mission lies in revisiting species sensitivity distribution (SSD) methods in the context of environmental risk assessment of pesticides on non-target terrestrial plants (NTTP). In particular, the applicant will have to:
- Select appropriate non-linear models to obtain statistically robust estimates of concentrations for which there is a 50% effect on the species tested (ER50);
- Define an SSD procedure to include censored ER50 estimates (ie defined as intervals and / or unbounded values) allowing to statistically estimate hazardous concentrations for 5% of species (HC5), covering the full range of sensitivities displayed in studied sets of NTTP;
- Apply this SSD procedure to construct an applicable decision tree for risk assessment based on case studies;
- Design a user-friendly user interface, compatible with good laboratory practices (GLP) applicable to computer systems, making it easy to implement the decision tree and the SSD procedure.


For NTTPs, multispecies-data are generated in line with the standard tests implemented in OECD guidelines 208 and 227. From these tests, ER50 values are estimated to be used as inputs within Species Sensitivity Distribution (SSD) analyses. Often noticed, some of these ER50 values are unbounded and nothing is clear within regulatory guidelines on how to deal with these values within SSD analyses. In order to lose such valuable data, the current SSD approach needs to be revisited, in particular for NTTPs; this is the main objective of the project in which the recruited person will be involved in order to: - Deliver guidance on how to implement the statistical analyses (ER50 / HC5 estimates) in a report; - Asses impacts on how the inclusion of unbound values will change the conservatism of HC5; - Built a decision tree when to include unbound values in ERA; - Establishing MOSAIC in the regulatory environment; This position will be co-supervised by the Laboratory of Biometry - Evolutionary Biology (resp. Sandrine CHARLES, CNRS, University Lyon 1, France) and the European Crop Protection Association (ECPA, resp. Virginie DUCROT et Peter SOWIG from Bayer Crop Sciences).


High-level skills in modelling and statistical inference, in R programming and if possible, in applied ecotoxicology, ecology and risk assessment. A deep knowledge of written and spoken English is absolutely necessary.

Work Context

The recruited person will work within the "Predictive Modeling and Ecotoxicology" team of the Laboratory of Biometry - Evolutionary Biology (CNRS UMR 5558 - University Claude Bernard Lyon I). The research in this laboratory is organised around two unifying themes: "Biometry" and "Evolutionary Biology". This specificity has presided over the creation of the laboratory and gives it its originality. The scientific results of the LBBE are in the synergy of methodological issues, whose core is mathematical modelling as well as the computer sciences, and also in the evolutionary perspectives which organises the researches independently of the level of biological organisation.

We are still looking for

We are still looking for applicants on our job offer for a research engineer / postdoc position in biostatistics applied to ecotoxicology (Lyon, France). The aim of the project is to revisit species sensitivity distribution (SSD) methods in the context of environmental risk assessment of pesticides on non-target terrestrial plants (NTTP). The duration is from 12 to 16 months.

Details are given here:

Applications must be done directly on-line.