Towards Development, Maintenance, and Standardized Phenotypic Characterization of Single‐Seed‐Descent Genetic Resources for Lupins

Well‐characterized genetic resources are fundamental to maintain and provide the various genotypes for pre‐breeding programs for the production of new cultivars (e.g., wild relatives, unimproved material, landraces). The aim of the current article is to provide protocols for the characterization of the genetic resources of two lupin crop species: the European Lupinus albus and the American Lupinus mutabilis. Intelligent nested collections of lupins derived from homozygous lines (single‐seed descent) are being developed, established, and exploited using cutting‐edge approaches for genotyping, phenotyping, data management, and data analysis within the INCREASE project (EU Horizon 2020). This will allow us to predict the phenotypic performance of genotyped lines, and will further boost research and development in lupins. Lupins stand out due to their high‐quality seed protein (∼40% of seed dry weight) and other primary components in the seeds, which include fatty acids, dietary fiber, and minerals. The potential of lupins as a crop is highlighted by the multiple benefits of plant‐based food in terms of food security, nutrition, human health, and sustainable production. The use of lupins in foods, along with other well‐studied and widely used food legumes, will also provide a greatly diversified plant‐based food palette to meet the Global Goals for Sustainable Development to improve people's lives by 2030. © 2021 The Authors. Current Protocols published by Wiley Periodicals LLC.

Lupins belong to the basally branching Genistoid clade of papilionoid legumes (Cardoso et al., 2013;Lavin, Herendeen, & Wojciechowski, 2005). Lupinus is a large and diverse genus within the legume family that comprises ∼280 species (Gladstones, 1998), although this number has been extended even further more recently, to up to 1000 species (Kurlovich, 2002). Lupin species are widespread across various climatic zones-which range from subarctic regions (e.g., Alaska), through the Mediterranean and the mountain regions of East Africa and Mexico as well as the Andes and the High Rockies, to subtropical regions of eastern South America-thus also highlighting their adaptation to varied agroclimatic conditions (Gladstones, 1988).
The lupin genus has been geographically separated into two centers of diversity. The species distributed throughout the Mediterranean basin and in North and East Africa form the group of 'Old World' lupins, while the species of North and South America constitute the 'New World' lupins (Gladstones, 1998). The Old World lupins include 13 to 15 species that are all annual and herbaceous, and these are mostly autogamous species (Gladstones, 1998;Pascual, 2004;Święcicki,Święcicki, & Wolko, 1996). In contrast, the majority of lupins are distributed throughout the New World. These species show diverse growth forms, which include annuals and herbaceous and woody perennials, with compound or unifoliolate leaves, and are well adapted to various ecosystems and climates across wide altitudinal ranges, e.g., coastal dunes, grasslands, and mountains (Aïnouche et al., 2004;Drummond, Eastwood, Miotto, & Hughes, 2012).
Whole-genome triplication appears to have been the main process involved in lupin evolution (Hane et al., 2016;Kroc, Koczyk, Swiecicki, Kilian, & Nelson, 2014;Xu et al., 2020), which has included multiple chromosome rearrangements (Susek et al., 2019;Susek, Bielski, Hasterok, Naganowska, & Wolko, 2016) and epigenetic changes (Susek et al., 2017). However, of particular note, the western New World lupins show an exceptionally high rate of diversification, with no evidence that polyploidy has had any role in the diversification of the species in this clade (Nevado, Atchison, Hughes, & Filatov, 2016). For this group of lupins, a rapid rate of speciation is believed to have been strongly associated with their transition from an annual to a perennial life history, along with their colonization of higher altitudes. This also enabled exploitation of novel ecological opportunities (Drummond et al., 2012;Nevado et al., 2016).
Old World and New World lupins were domesticated independently, with the incorporation of key traits of the domestication syndrome, such as non-shattering pods, permeable seed coats, and large seeds . Lupinus albus (white lupin; annual, 2n=50; genome size, ∼580 Mbp) dates back to the times of the ancient Greeks and Romans. At that time, in 1000-800 BC, farmers used white lupin for soil improvement and crop rotation while also using the seeds for food and animal feed. They were the first who selected for large permeable seeds and non-shattering pods. Greece and the Balkan Peninsula are considered to have the greatest diversity of L. albus and its wild subspecies (subsp. graecus, subsp. termis, subsp. albus) and cultivated types (i.e., landraces) (Gladstones, 1998). Lupinus mutabilis (tarwi, Andean lupin, pearl lupin; annual, 2n=48; genome size, ∼930 Mbp) is considered to have been primarily domesticated in the Andes between 1800 and 2600 BC . The latest data suggest that tarwi was domesticated in the highlands of northern Peru (most likely the Cajamarca region), and that Lupinus piurensis is the likely progenitor of tarwi . The history of tarwi domestication inferred from demographic analyses also suggests that after the split from its progenitor, it went through a classical domestication bottleneck, with subsequent rapid population expansion as a widely cultivated species across the Andes . Nowadays, L. mutabilis is only known in cultivation, with no wild populations discovered to date. The South American Indians used L. mutabilis as food (after de-bittering), and as green manure and medicine (e.g., for cardiac disease, rheumatism, malaria) (Bebeli et al., 2020). Interestingly, the seed size increased two-fold (i.e., doubled) during the domestication of L. mutabilis .
Lupinus albus and L. mutabilis have therefore been traditional food legumes in the Mediterranean regions and the Andes, respectively, for thousands of years (Cowling, Buirchell, & Tapia, 1998). The seeds of L. albus and L. mutabilis are relatively large, and across lupins in general, they are characterized by the highest protein content (up to 38%, 41%-51% of seed dry weight, respectively) and oil content (9%-13%, 14%-24% of seed dry weight, respectively), and can thus be used as dual-purpose, providing both protein and oil for human consumption (Gresta et al., 2017;Gulisano, Alves, Martins, & Trindade, 2019). Moreover, the oil from L. albus and L. mutabilis seeds is of high quality, Kroc et al.

of 21
Current Protocols with a high proportion of unsaturated fatty acids, low erucic acid, and a long shelf life (Cowling et al., 1998;Rybiński et al., 2018). The main anti-nutritional factor of lupin seeds is the alkaloid content (Wink et al., 2010). However, owing to intensive breeding efforts, the seed alkaloid contents of modern lupin cultivars are often lower than the accepted industry threshold (Kamel,Święcicki, Kaczmarek, & Barzyk, 2016;Kroc et al., 2017).

LUPIN GENOMICS AND GERMPLASM DIVERSITY
Lupin genetics, genomics, and germplasm resources lag far behind those of other major crops. Substantial progress has been made recently in the molecular characterization of the L. albus transcriptome (Secco, Shou, Whelan, & Berkowitz, 2014;Wang et al., 2014), with two genome assemblies of cultivar 'Amiga' (Hufnagel et al., 2020a;Xu et al., 2020) and a pangenome assembly (Hufnagel et al., 2020b). These provide basic resources for studies into the biology and breeding of this species, and will allow future investigations into the influence of white lupin domestication on its genomic variability (Hufnagel et al., 2020b). On the other hand, L. mutabilis remains an under-studied crop, which still lacks extensive genomics resources. Recently, a large genome-wide DNA polymorphism dataset was generated for L. mutabilis using 'next restriction site associated DNA sequencing' (nextRADseq; Atchison et al., 2016). L. mutabilis was also included in a chloroplast genome investigation that provided comparative analysis and further resolution of phylogenies within Lupinus (Keller et al., 2017). The genetic diversity within L. mutabilis and its relation to other lupin species was illustrated with aid of molecular markers (Chirinos-Arias, Jiménez, & Vilca-Machaca, 2015; Olczak, Rurek, Jańska, Augustyniak, & Sawicka-Sienkiewicz, 2001;Talhinhas, Neves-Martins, & Leitao, 2003).
Data on the variations within the genetic resources for white lupin and tarwi are scarce. A set of target phenotypic traits and phenology showed variations in both wild and domesticated white lupins that were grouped into seven clusters (Berger, Shrestha, & Ludwig, 2017). Here, the number of days to flowering varied from 65 days for an average Mediterranean climate (cluster 5) to 70 days for the cooler long-season Iberian climate with higher rainfall (cluster 2). White lupins have been defined in the context of large seeds, high early vigor, rapid growth rates, plant heights, harvest indices, and seed and biological yields (Berger et al., 2017), along with their characterization according to differences in the quality of their seed contents.
A valuable source of Ethiopian white lupin variations has been described, where about 500 white lupin genotypes were collected and are conserved at the Ethiopia Biodiversity Institute (Beyene, 2020). Molecular analyses of 212 Ethiopian landraces revealed the high genetic diversity, and highlighted distinct gene pools (Atnaf et al., 2017). A local, high-alkaloid, white lupin variety has also been described that is grown in north-western Ethiopia, which is also partially non-shattering, high-yielding, and most importantly, resistant to lupin anthracnose disease (Yeheyis, Kijora, Melaku, Girma, & Peters, 2010). In addition, 25 landraces in north-western and southern Ethiopia were characterized by their differences in days after sowing for emergence time, first flowering, 50% flowering, and maturity, along with plant height (Beyene, 2020). Variations in landraces of white lupin have also been reported to include their yield characteristics, as numbers of pods per plant, numbers of seeds per pod, and pod length, and also in terms of their protein Kroc et al.

of 21
Current Protocols content. Some other seed traits are not significantly different across these Ethiopian landraces (e.g., 100-seed weight, seed length, and width; Beyene, 2020).
The genetic diversity of L. mutabilis is shown by the phenotypic variations in its flower, stem, and seed colors, and its indeterminate growth. The variations in seed traits relate to their shape (from lenticulate to spherical) and primary color (from pearly white to dark, with some intermediates), along with various patterns of pigmentation. However, most L. mutabilis cultivar accessions of Andean germplasm collections have a pearl white color (i.e., 95%). Lupins with a dark seed color also have darker flowers, which suggests that the white color is recessive (Gulisano et al., 2019). Seed protein and oil content were shown to vary among 149 Peruvian tarwi accessions, which suggested that the seeds of the taller and larger plants tended to be richer in protein, while those of the smaller plants were richer in oil (Neves Martins, Talhinhas, & De Sousa, 2016). Of note, an analysis of the variations across 23 L. mutabilis accessions in terms of their combined genetic and genomic characteristics and morphological traits revealed variations in many traits, such as time to flowering, plant height, seed weight, and number of pods (Guilengue, Alves, Talhinhas, & Neves-Martins, 2020). It has been suggested that indeterminate and determinate plants are adaptive to interannual meteorological variations under southern (Mediterranean) and northern and central European conditions, respectively (Neves Martins et al., 2016).
The white and tarwi lupin germplasm collections that are conserved in genebanks across the world generally include both cultivated material and wild populations. Nonetheless, these materials have not been well characterized. Most accessions indicate the country of origin only, with the other 'passport' data limited or missing. According to the European Search Catalogue for Plant Genetic Resources database (EURISCO Catalogue; http:// eurisco.ecpgr.org, 2020-12-02), the greatest number of white lupin genetic resources are conserved in Portugal (927 accessions For L. mutabilis, the most numerous genetic resources in Europe are conserved in Germany (695 accessions), Portugal (150), and the Russian Federation (132) (http: // eurisco.ecpgr.org, 2020-12-02). The largest and most relevant germplasm collections of tarwi outside Europe are held mainly in the U.S.A. (102) and Australia (74) (https: // www.genesys-pgr.org, 2020-12-02), and also Peru, Ecuador, and Bolivia. However, reports suggest that much of the L. mutabilis diversity remains uncollected (Gulisano et al., 2019;Jacobsen & Mujica, 2008), and little is known about the genetic variability of these collections (Guilengue et al., 2020).

DEVELOPMENT AND MAINTENANCE OF THE LUPIN INTELLIGENT COLLECTION: INCREASE
The need to meet the global challenges for the management and use of genetics resources was very recently, and urgently, emphasized by Mccouch et al. (2020). They highlighted that joint efforts are crucial for efficient and expeditious characterization and use of agrodiversity, through the establishment of the necessary platforms to empower genebank managers, researchers, breeders, and farmers to more effectively use genetic variation not just for research but also for accelerated crop improvement and sustainable production. The progress and shift in research strategies from traditional plant phenotyping to combined phenotypic and genotypic studies (e.g., association studies, using the whole genome) now pave the way for the accurate and comprehensive definition of this diversity.
Kroc et al.

of 21
Current Protocols To date, there has been limited information available on lupin phenotypic descriptors, and even when available, the data have generally only provided basic descriptive information. The exception here is the Agricultural Research Service of the U.S. Department of Agriculture, where some images of seeds are already included in the collection, although without any protocol to define how to image and analyze the seed traits (https:// npgsweb.ars-grin.gov/ gringlobal/ taxon/ taxonomydetail?id=22802). According to the International Treaty on Plant Genetic Resources for Food and Agriculture, only 644 white lupin and 706 tarwi accessions have digital object identifiers (DOIs), with these mainly registered by genebanks, e.g., the Leibniz Institute of Plant Genetics and Crop Plant Research in Germany (https:// ssl.fao.org/ glis).
The INCREASE project gets its name from "Intelligent Collections of Food Legume Genetic Resources for European Agrofood Systems," and is funded by the European Union as part of Horizon 2020 (Bellucci et al., 2021). This has provided the unique opportunity to develop phenotyping data and to integrate these with genotypic information for thousands of lupin accessions from different sources (e.g., genebanks, research institutes, private companies, project stakeholders). The protocols for exploring the variations within lupins and other food legumes studied within INCREASE (i.e., also chickpea, common bean, lentil; see Current Protocols articles: Guerra-García, Gioia, Von Wettberg, Logozzo, & Bett, in preparation; Kumar et al., in preparation;and Cortinovis et al., 2021) are being provided to enhance the standardization and data handling of the available genetic resources of these legumes and to increase the effectiveness of their conservation and use (https:// www.pulsesincrease.eu).
Thus, a set of lupin nested core collections is being established within the INCREASE project, as the Reference-core (R-core), Training-core (T-core), and Hyper-core (H-core) collections. These will be multiplied, characterized genetically and/or phenotypically, and conserved for further use. The wide range of phenotypic traits will be defined and analyzed to provide the set of genotypes that show the phenotypic diversity, to group genotypes according to their similarities and dissimilarities, and to integrate these with various types of genetic and genomic data.
In the present study, we describe the procedures that are established and implemented in the INCREASE project to develop lupin Intelligent Collections. These protocols are also recommended for use in genebanks and research institutions and are aimed at the development of single-seed descent (SSD) lines for conservation and maintenance of seeds. The protocols will facilitate the characterization of lupin genetic resources and the integration of the data obtained into both centralized and decentralized systems, which will eventually be accessible to each end user. We have developed the basic protocols for the lupin primary seed increase based on Lupin Descriptors ( Cortinovis et al. (2021). The scheme for the development of the SSD lines and seed multiplication cycles, as well as for the sharing and management of the genetic resources within the IN-CREASE project, is illustrated in Figure 1. This workflow includes the following steps: • Step 1: List of genetic resources. The basic initial step is the preparation of the list of accessions obtained from various providers, and to order these with their unique INCREASE codes, so as to be able to manage the accessions properly and avoid any mistakes while sharing seeds. Moreover, DOIs will be assigned to the accessions that are selected for the R-core, by registering them on the Global Information System of Kroc et al.

of 21
Current Protocols Figure 1 Scheme for the development of the single-seed descent lines (SSD) and seed multiplication cycles, as well as for the sharing and management of the genetic resources.
the International Treaty on Plant Genetic Resources for Food and Agriculture (https:// ssl.fao.org/ glis). It is essential to share the seeds under easy Standard Material Transfer Agreements or policies. • Step 2: Characteristics of the seeds from heterogeneous accessions. The seeds indicated for the development the SSD lines are characterized according to Basic Protocol 1 (Lupin seed phenotypic descriptors), with their image analysis using Basic Protocol 2 (Lupin seed imaging).
Of importance: Pictures of the original seeds (heterogeneous material) used to develop these SSD lines, as well as the SSD seeds at the subsequent selfing cycles, should be archived for further comparison between the originals and the next generations of SSD genotypes. • Step 3: Development of SSD lines. One single seed of each heterogeneous accession is chosen at random to be grown under insect-free conditions (i.e., greenhouse). Recommended: Seeds are surface-sterilized to avoid potential development of pathogen-caused diseases, and are cold-treated for 21 days at 6°to 8°C to induce flowering in genotypes responsive to vernalization. Of importance: The phenotypic observations for each genotype are carried out for each cycle of the SSD line development to compare the phenotypic traits among the SSD generations, using Basic Protocol 3 (Standardized phenotypic characterization of lupin genetic resources grown for primary seed increase-development of SSD genetic resources). • Step 4: The multiplication cycles. One single seed from the first SSD line is chosen at random to develop the second selfing cycle, according to Step 3, above (same approach should apply in the subsequent selfing cycle; e.g., to produce a third cycle of SSD lines, the single seed from the second round of selfing is chosen at random). Seed characterization and imaging are carried out as described for Step 2. • Step 5: SSD line sharing and conservation. The SSD lines developed are shared among different genebanks (and stakeholders) to multiply the seeds themselves, using the INCREASE lupin protocols, and to integrate and compare the data provided in the reciprocal approaches. • Step 6: Data integration. A large amount of phenotypic information from each genotype for each particular seed selfing cycle will be obtained. These data will be integrated with large-scale genotypic data into the INCREASE information technology system, so that the data can be stored, shared, and explored by multiple users through an open and publicly available web portal.
Kroc et al.

of 21
Current Protocols

LUPIN SEED PHENOTYPIC DESCRIPTORS
This protocol describes the assessment of the seed morphological traits prior to each seed multiplication cycle. The seed traits are recorded at the beginning of each seed increase cycle, starting from the first cycle of SSD line development using the heterogeneous genotypes. These assessments will enable comprehensive characterization of seeds from different lupin genetic resources, as well as detection and elimination of errors that might occur during the SSD cycles.

Materials
Seeds of lupin genetic resources Validated measuring devices, such as a ruler A standardized color scale (i.e., ColorChecker Classic Nano, X-Rite) Template file (spreadsheet)

NOTE:
The seed traits listed below should be recorded at the beginning of each primary seed multiplication cycle.
1. Take at least five seeds from each accession (heterogeneous materials or SSD lines), and through visual assessment record the Seed Shape, according to the following categories ( Fig. 2): 1 = spherical 2 = flattened spherical (lenticular) 3 = oval 4 = flattened oval 5 = cuboid 6 = flattened cuboid 2. Take at least five seeds from each accession (heterogeneous materials or SSD lines), and through visual assessment record the Seed Primary Color, according to the following categories: The seed primary color is the dominating basal color of the seed.  3. Take at least five seeds from each accession (heterogeneous materials or SSD lines), and through visual assessment record the Seed Ornamentation, according to the following categories (for example of lupin seed ornamentation, see Fig. 3): 1 = crescent 2 = eyebrow 3 = back 4 = spotted 5 = moustache 6 = marbled 7 = marbled crescent 8 = marbled plus eyebrow 9 = spotted plus eyebrow 10 = spotted plus moustache 11 = other Ornamentation refers to well-defined seed coat pattern that is different from the primary color. It should be assessed at full maturity of the seeds.
4. Take at least five seeds from each accession (heterogeneous materials or SSD lines), and through visual assessment record the Seed Ornamentation Color, according to the following categories: 1 = gray 2 = light brown 3 = dark brown 4 = black 5 = other Density of seed ornamentation 5. Take at least five seeds from each accession (heterogeneous materials or SSD lines), and through visual assessment record the Density of Seed Ornamentation, according to the following categories (Fig. 4): 1 = sparse 2 = medium 3 = dense 4 = very dense 6. Take at least five seeds from each accession (heterogeneous materials or SSD lines), and through visual assessment record the Seed Quality, i.e., visibly healthy seeds, not affected by any pest or disease.

LUPIN SEED IMAGING
This protocol describes the best-practice standards for seed imaging within the IN-CREASE project. The images of the seeds are archived prior to each seed multiplication cycle, starting from the first cycle of SSD line development using the heterogeneous genotypes. Such records will serve as a documentation of seed quality, as well as shape, color, and size of seeds, and might be an input for automated image analysis.

Materials
Seeds of lupin genetic resources Validated measuring devices, such as a ruler A standardized color scale (i.e., ColorChecker Classic Nano, X-Rite) Uniform background Digital camera (≥20 MP resolution) 1. Take at least five seeds from each accession (heterogeneous materials or SSD line).
Seeds representing the entire variation in seed morphological characteristics should be chosen (i.e., color or shape).

Place the seeds on the uniform background.
Preferably, a color not observed on seeds should be chosen, e.g., light gray.
3. Place the seeds close to the ruler.

of 21
Current Protocols 5. Place the standardized color scale to provide color quality control and enable automated image post-processing.
The bioinformatics group of the INCREASE project is working on software using artificial intelligence (AI) approaches for this.

Add accession label or leave enough space for the subsequent insertion of the label information and the barcode (QR code).
Images should be clearly labeled including unique IncreaseID followed by the sample ID. Full Image-ID description is structured as follows: <IncreaseID>_ <SampleID>_<ImageID> 7. Take a picture of each accession.
The resolution and size of an image depends on the intended use; e.g., resolution for research, printing, and web presentation should be 600 dpi, 300 dpi, and 72 dpi, respectively. The color bit depth should be 24-bit color.
8. Transfer the finalized images to a central repository for the project.

STANDARDIZED PHENOTYPIC CHARACTERIZATION OF LUPIN GENETIC RESOURCES GROWN FOR PRIMARY SEED INCREASE (DEVELOPMENT OF SSD GENETIC RESOURCES)
Here we describe the protocol to phenotypically characterize the lupin genetic material grown under controlled conditions, for primary seed increase. The protocol should be applied at the stage of SSD development from heterogeneous material, as well as for further selfing cycles. Collected data will be uploaded into the database of the project and integrated with the results obtained for the same line in other experiments (e.g., field trials, molecular characterization), thus serving as a comprehensive information resource.

L. albus and L. mutabilis plants grown in insect-free conditions
The following conditions are offered for guidance in providing optimal growing conditions for lupins: pod capacity, 7.5 L; potting medium, peat and vermiculite mixture (1:1 by volume); day and night temperatures of 22°C and 18°C, respectively; at last 14-to 16-hr photoperiod; air humidity, 60%-65%; plant watering as required, average, 200 ml per pot, twice a week; the commercial rhizobia inoculum treatment might be applied prior to sowing (to the seeds or soil); during the flowering period plants to be fertilized once a week; biological pest controls as well as protective biological products (e.g., mites, microorganisms) to be used through the whole experimental period; chemical pest control to be used when other attempts have failed. Validated measuring devices, such as a meter stick or a ruler Data collection template file (spreadsheet) NOTE: Prior to sowing, it is recommended to surface-sterilize the seeds to avoid potential development of pathogen diseases. As this approach weakens the seed coat, it might serve as an scarification treatment to stimulate germination. Soak the seeds in 70% ethanol (30 s for L. albus, 20 s for L. mutabilis); rinse the seeds under tap water; soak the seeds in 1.5% sodium hypochlorite (4 min for L. albus, 1 min for L. mutabilis); rinse the seeds under sterile water; and put the seeds into sterile petri dishes and then in the refrigerator.
1. Prior to protocol implementation, collect the following information on the experimental site: Insect-free conditions: greenhouse/tunnel/ growth chamber Growing conditions (day/night temperature, photoperiod, humidity, watering) Kroc et al.

of 21
Current Protocols

Record the Dates of Cold Treatment (dd/mm/yyyy).
Exposing the seed to cold temperature (6-8°C) for 21 days induces flowering; the time of cold treatment might be adjusted to the individual genotype.

Current Protocols
An open flower is defined as one with visible wings and standard petal (Fig. 6C). 1 = white 2 = white/blue; white color is more expanded than blue 3 = blue 4 = blue/white; blue color is more expanded than white 5 = white/violet; white color is more expanded than violet 6 = violet 7 = violet/white; violet color is more expanded than white 8 = white/pink; white color is more expanded than pink 9 = pink 10 = pink/white; pink color is more expanded than white 12. Through visual assessment, record the dominant Color of Standard Petal in the Just-Opened Flower of the Main Stem, according to the following categories: 1 = white 2 = white/blue; white color is more expanded than blue 3 = blue 4 = blue/white; blue color is more expanded than white 5 = white/violet; white color is more expanded than violet 6 = violet 7 = violet/white; violet color is more expanded than white 8 = white/pink; white color is more expanded than pink 9 = pink 10 = pink/white; pink color is more expanded than white Withering is defined as when the first petal dies and become brownish ( Fig. 7A and B).

of 21
Current Protocols 21. Through visual assessment, record the Purple Pigmentation of the Main Stem at 2 months and 4 months after emergence, according to the following categories: 1 = none 2 = slightly 3 = medium intensity-more than half of the stem is purple 4 = very intense-stem and internodes are purple 5 = no pigment in stem plus purple internodes 22. At flower withering, record the Growth Habit, according to the following categories ( Fig. 8

Record the Date of First Pod Set on the Main Stem (dd/mm/yyyy).
Pod set = the pod is 8 to 10 mm long ( Fig. 9A and B).  Pod set = the pod is 8 to 10 mm long ( Fig. 9A and B).

Record the Date of First Pod Color Changes on the Main Stem (dd/mm/yyyy).
The pod color has changed from green to khaki. These pods are ready to be harvested soon ( Fig. 9C and D).

Record the Date of First Pod Color Changes on the Lateral Branches (dd/mm/yyyy).
The pod color has changed from green to khaki. These pods are ready to be harvested soon. When pods have dried, the seeds in the pods have reached full size and the pods can no longer be dented with the thumbnail (Fig. 9E).
30. Record the Date of Harvesting Pods on the Lateral Branches (dd/mm/yyyy), i.e., when the pods have dried and turned a golden-brown color.
When pods have dried, the seeds in the pods have reached full size and the pods can no longer be dented with the thumbnail (Fig. 9E). The pod and seed growth phase begins when a pod is set (reaches 8-10 mm) and ends when the pod changes color (from green, through khaki, to light brown).

Conclusions and Future Prospects
Comprehensive characterization, management, and use of the genetic resources is crucial for enhancing lupin agrobiodiversity and efficient exploitation of these species. Worldwide germplasm collections of lupin, which is one of the most important food legumes, are not well characterized. Within the INCREASE project, intelligent nested collections of the white and tarwi lupins will be established and exploited using cuttingedge approaches to genotyping, phenotyping, and data management and sharing. Here, we provide ready-to-use protocols to develop and describe SSD lines, comprising the broad set of phenotypic traits that can be used. These protocols will facilitate the screening and maintenance of lupin germplasm variations for agricultural improvement.
These protocols are valuable templates for a wide range of users, such as genebank managers, researchers, breeders, and farmers. Their use will not only reduce commonly seen misunderstandings, but more importantly will provide the gold standard for lupin genetic resource characterization and use when Kroc et al.

of 21
Current Protocols targeting specific traits. The approach to germplasm characterization and management that is being developed within the INCREASE project represents a fundamental basis for integration of the increasingly accessible genomic resources, along with the advanced 'omics' analyses, to accelerate the discovery of improved lupin agricultural diversity and to determine valuable sources for breeding programs.