Discrimination between regional biotypes of Impatiens glandulifera using a simple MALDI-TOF MS-based method for use with seeds

Table 2 Average Bruker score against triplicate reference spectra from seeds collected in 2017 for 12 blind-test seeds collected in 2016 from Harmondsworth Moor, Lampeter, Rhosmaen, and Silwood Park (three seeds from each site)

Blind-test number	Average Bruker score against triplicate reference spectra for seeds from				Identification call	Accuracy after unblinding
Blind-test number	Harmondsworth Moor	Lampeter	Rhosmaen	Silwood Park	Identification call	Accuracy after unblinding
13	1.455	2.170	2.115	1.842	Lampeter	Correct
14	0.972	2.247	2.165	1.949	Lampeter	Incorrect (Rhosmaen)
15	2.322	1.390	1.378	1.251	Harmondsworth Moor	Correct
16	1.155	2.014	2.026	1.639	Rhosmaen	Correct
17	2.041	1.525	1.487	1.603	Harmondsworth Moor	Correct
18	1.384	1.433	1.650	2.399	Silwood Park	Correct
19	1.325	2.207	2.181	1.645	Lampeter	Correct
20	1.404	1.495	1.621	2.336	Silwood Park	Correct
21	1.710	1.855	1.996	2.350	Silwood Park	Correct
22	1.136	2.185	1.936	1.772	Lampeter	Correct
23	2.006	1.754	1.734	1.587	Harmondsworth Moor	Correct
24	1.488	2.280	2.371	1.779	Rhosmaen	Correct

The highest average Bruker score is given as the identification call in the penultimate column, and the accuracy of this after unblinding the data is shown the final column

ISSN: 1746-4811