Table 3. Summary of proteomics data.

	Protein discovery (PD)				Highest quality (HQ)
Statistic	O. exa.	O. gar.	O. sph.	Total	O. exa.	O. gar.	O. sph.	Total
N raw spectra	9676	8419	7840	25 935	7433	6648	6344	20 425
N clean spectra	9357	8181	7588	25 126	7127	6426	6102	19 655
N unique peptides	4072	3835	4137	7487	2932	2878	3164	5496
N proteins (transcriptome)1	1748	1672	1610	2644	1201	1222	1206	1603
N proteins (non-transcriptome)2	403	405	439	887	130	134	143	204
Total N proteins	2110	1943	2031	3531	1331	1356	1349	1807

Data are presented for species individually and for the total of all three species combined. The protein discovery (PD) analysis was performed at a protein identification threshold of 90%, the highest quality (HQ) data set was compiled at a threshold of 99%. For spectra, numbers are given before (‘raw’) and after (‘clean’) removal of known contaminants. All other numbers were obtained from cleaned spectra only.

¹Proteins matching a sequence in the Ophrys reference transcriptome. In the HQ data set, 115 were 454 singleton reads, 4 Sanger singleton reads, 388 Solexa contigs, 343 pyrosequencing isotigs and 753 combined 454/Solexa contigs. Among these 1603 HQ orchid proteins, no peptide matched translations from different strands, and only 11 proteins (0.7%) had peptides matching to two reading frames of the same transcript strand (in each case due to a single frame shift).

²Proteins not matching any Ophrys transcript, but with SwissProt and/or TAIR9 database hit.