35
SEED TESTING INTERNATIONAL APRIL 2026
• RULES DEVELOPMENT
markers tested in CT1. The markers were tested 
as three different marker systems/panels: one 
had four markers and the other two each had 
five markers. Each participant received ten 
tubes with subsamples and two tubes of pools 
of 40 crushed seeds per variety. Samples were 
provided in sealed tubes labelled with the name 
or code of the variety. Duplicates were provided 
to all participants as a backup. However, if any 
participating laboratory ran out of a sample, they 
were able to contact the CT leader to request 
more material.
Equipment, Chemicals and Procedure 
Inclusion of DNA-based methods into the ISTA 
Rules is semi-performance-based. Laboratories 
were provided with guidelines for running the 
SSR prescribed, but the specific procedure was 
ultimately up to the participating laboratories.
Evaluation and Reporting of Results 
Results were reported in an Excel spreadsheet 
indicating laboratory number, variety name, SSR 
name and allele sizes. Five of the six laboratories 
that participated in CT1 sent a data package. 
The data analysis from CT1 aimed to evaluate 
if the marker panel was reproducible among 
laboratories and thus suitable for being kept for 
CT2, and eventually for the ISTA Rules proposal. 
This evaluation was carried out by the crop leader 
and consisted of verifying if markers gave the 
same allele pattern across laboratories (even if 
different equipment and reagents were used). 
For CT2, the group leader compiled the results 
and prepared an Excel file with allele sizes and 
binary data. Binary data was sent to the Chair of 
the ISTA Statistics Committee for their analysis. 
All laboratories that participated in CT2 sent a 
data package.
Statistical Analysis 
Overall percentage agreements (pα) and Cohen’s 
kappas have been computed for all the possible 
laboratory pairs, considering as units either the 
marker alleles or the varieties. The computations 
have been performed with the R irr package 
(Gamer et al., 2012), which includes functions 
for computing various coefficients of reliability 
of agreement. Within-laboratory agreement has 
been assessed through accordance as described 
in Langton et al. (2002). The average accordance 
across varieties is high in all the laboratories 
(above 95%) except in laboratory G (92.3%, when 
considering only individual seeds). Agreement of 
the marker results across laboratories has been 
assessed with Fleiss’ kappa κ on all the varieties 
except varieties 3 and 4. The agreement is perfect 
(κ=1) for all the varieties according to the Landis 
and Koch (1977) table for the interpretation of κ. 
The conclusion of the statistical analysis is that 
given these results, there is enough evidence for 
validating the method.
Final Comments and 
Conclusions
After running two CTs for barley varietal 
identification using a panel of 14 SSR markers 
and with the participation of ten laboratories 
from around the world, the statistical analysis 
done by the ISTA Statistics Committee concluded 
that there is enough evidence for validating the 
method for four SSR markers. Given the work 
carried out and the conclusion of the Statistics 
Committee, the ISTA Variety Committee presents 
this validation report for considering the 
inclusion of the barley SSR marker panel in the 
ISTA Rules, Chapter 8.
References
1. Gamer, M., Lemon, J., Fellows, I. and  
Singh, P. (2012). Package ‘irr’: Various  
Coefficients of Interrater Reliability and 
Agreement. The Comprehensive R Archive 
Network, Vienna.
2. ISTA (2025). International Rules for Seed 
Testing. International Seed Testing Association, 
Wallisellen, Switzerland.
3. Langton, S.D., Chevennement, R., 
Nagelkerke, N. and Lombard, B. (2002). 
Analysing collaborative trials for qualitative 
microbiological methods: accordance and 
concordance. International Journal of Food 
Microbiology, 79, 175–181.
4. Landis, J.R. and Koch, G.G. (1977). The 
measurement of observer agreement for 
categorical data. Biometrics, 33, 159–174.
5. Perry, D.J., Fernando, U. and Lee, S.-J. (2013). 
Simple sequence repeat-based identification  
of Canadian malting barley varieties.  
Canadian Journal of Plant Science, 94, 485–496.

View this content as a flipbook by clicking here.