In this study, a total of 10 high-PA-content cash crops with currently available genomic databases were selected: almond (Prunus dulcis), apple (Malus domestica), blueberry (Vaccinium corymbosum), cacao bean (Theobroma cacao), common bean (Phaseolus vulgaris), grape (Vitis vinifera), peanut (Arachis hypogaea), soybean (Glycine max), strawberry (Fragaria × ananassa), and tea tree (Camellia sinensis) (Table 1). These crops play a critical role in the global food market and cosmetic industry; according to statistical data from the Food and Agriculture Organization of the United Nations (www.fao.org, accessed on 5 February 2021), apples, grapes, and soybeans were ranked in the top 20 for commodity production in the USA in 2019. Almonds and blueberries were also produced at capacities of about 193.7 kt and 30.9 kt, respectively, in 2019 in the USA. Peanuts and strawberries are important commercial crops in several countries, including China, India, Indonesia, and the USA. In China, which has the largest market, the trade volumes of peanuts and strawberries were USD 471 million and 75 million, respectively, in 2019. Cacao beans are consumed worldwide as an ingredient in a variety of processed products—such as chocolate, cocoa powder, and cocoa butter—at a capacity of 3991 kt (in 2016, with the latest available data), as reported by Statista (www.statista.com/, accessed on 6 February 2021).

The PA content ranged from 145.0 to 3532.2 mg/100 g among the 10 species (Table 1). Grape seeds have been reported to have the highest level of PAs (3532.2 g/100 g), followed by cacao and common beans (1460.0 g/100 g and 1000.1 g/100 g, respectively) [2,14,34]. The levels of PAs detected in the 10 species can vary depending on the cultivars or measuring conditions used in each study, and the amount of actual absorption by humans can change depending on the manner of intake as food; for example, fresh grapes have much lower levels of PAs (0.05 g/100 g) than grape seeds (3532.2 g/100 g) [2]; in soybeans, seed coats contained significantly higher contents of PAs than the embryos [38]. In common beans, the levels of anthocyanin and PAs vary according to the color features of the bean coats, depending on the genotypes [34,85].

Species	Proanthocyanidins Content (mg/100 g)	Reference	Genome Database	Reference	Assembly Size (Mb)	Coverage (%)	Contig N50 (Kb)	Number of Genes Predicted
Almond (Prunus dulcis)	184	Prior & Gu, 2005 [14]	Prunus dulcis Lauranne Genome v1.0 (http://rosaceae.org/, accessed on 12 February 2021)	Alioto et al., 2020 [31]	227	95	103	27,969
Apple tree (Malus domestica)	162	Hellström et al., 2009 [2]	(iris.angers.inra.fr/gddh13/, accessed on 12 February 2021)	Daccord et al., 2017 [32]	643	100	620	42,140
Blueberry (Vaccinium corymbosum)	255	Prior & Gu, 2005 [14]	V_corymbosum v1.0 (http://gigadb.org/, accessed on 12 February 2021)	Colle et al., 2019 [33]	1680	102	15	32,140
Cacao bean (Theobroma cacao)	1460	Hellström et al., 2009 [2]	Cacao Matina1-6 Genome v2.1 (http://cacaogenomedb.org, accessed on 12 February 2021)	Publication in progress (http://cacaogenomedb.org)	346	80	1080	27,379
Common bean (Phaseolus vulgaris)	1000	Kan et al., 2016 [34]	Phaseolus vulgaris v2.1 (http://phytozome.jgi.doe.gov/, accessed on 12 February 2021)	Schmutz et al., 2014 [35]	600	80	1900	27,433
Grape (Vitis vinifera)	3532	Prior & Gu, 2005 [14]	Vitis vinifera v2.1 (http://phytozome.jgi.doe.gov/, accessed on 12 February 2021)	Jaillon et al., 2007 [36]	487	102	566	26,346
Peanut (Arachis hypogaea)	186	Hellström et al., 2009 [2]	(http://peanutgr.fafu.edu.cn/, accessed on 12 February 2021)	Zhuang et al., 2019 [37]	2538	94	1509	83,709
Soybean (Glycine max)	300	Lee et al., 2017 [38]	Glycine max Wm82.a4 (http://www.soybase.org, accessed on 12 February 2021)	Schmutz et al., 2010 [39]	1150	95	1492	46,430
Strawberry (Fragaria × ananassa)	145	Prior & Gu, 2005 [14]	(https://datadryad.org/, accessed on 12 February 2021)	Edger et al., 2019 [40]	813 *	99	79	108,087
Tea tree (Camellia sinensis)	189	Engelhardt et al., 2003 [12]	(http://tpia.teaplant.org/, accessed on 12 February 2021)	Xia et al., 2019 [41]	2890	95	67	53,512

Despite the nutritional importance of flavonoids, key enzymes involved in these pathways have not been identified in many major food crops with high flavonoid contents (Table 1). To characterize the PA biosynthetic pathways in the selected species, the orthologous genes encoding key enzymes involved in all pathways were searched in the latest reference genomic databases of each species (Table 1). N50 and genome coverage are two of the most important factors for evaluating the quality of genome assembly, where N50 is defined as the length of the contig, scaffold, super-scaffold, and pseudomolecule together being shorter than or equal to 50% of the total genome assembly length, and coverage is calculated from the percentage of the total assembly size over the reference genome. In this study, contig N50 varied widely, ranging from 79 kb (strawberry) to 1900 kb (common bean), and the highest, lowest, and mean genome coverage were 102% (blueberry and grape), 80% (common bean and cacao), and 94%, respectively (Table 1).

To identify orthologous genes involved in the PA pathway in the major crops, we collected a list of key enzymes (F3′5′H, F3′H, F3H, DFR, LAR, ANS, and ANR) based on the EC number from the KEGG database. Of the eight species, 92 genes were identified, including two wild relative species of strawberry and peanut (Arachis duranensis: 6, A. thaliana: 6, Fragaria vesca: 4, G. max: 20, M. domestica: 12, P. vulgaris: 12, T. cacao: 9, V. vinifera: 23) (Table 2). The protein sequences were downloaded from the NCBI GenBank database (www.ncbi.nlm.nih.gov, accessed on 4 March 2021) and cross-checked using the reference genomic database of each species (Table 2). We found that the sequences of three genes—XP_014632702.1 (G. max), XP_007138248 (P. vulgaris), and XP_017985307.1 (T. cacao)—were not available in the latest reference annotation data (Supplementary Table S1). In total, 83 genes were identified in the reference genome, and 14 paralogs, which had not been reported to be involved in the pathways in the KEGG database, were additionally detected in two species (A. hypogea: 5 and F. ananassa: 9; Table 2). Using these 97 genes, orthologous genes were identified in the eight crop species, and 105 genes were newly identified as candidate genes involved in the flavonoid biosynthetic pathway from five species (A. hypogea: 1, C. sinensis: 19, F. ananassa: 7, P. dulcis: 4, and V. corymbosum: 74) (Table 2, Supplementary Table S2). The annotation data currently available for the newly identified orthologous and paralogous genes agreed well with the EC numbers from the KEGG database. In C. sinensis, P. dulcis, and V. corymbosum, a total of 97 orthologs were first found in our study, and they have not been reported to be involved in the phenylpropanoid pathway according to the KEGG pathway.