Allium L. falls under one of the widespread and main genera in the Amaryllidaceae family (Friesen et al. 2006; Li et al. 2010; Fritsch et al. 2010). As of Linnaeus, the number of species have risen from 30 to over 920 species at present (Govaerts et al. 2005-2014). There are some species of economic importance in the genus, namely onion, garlic, leek, shallot, bunching onion, and chives planted as vegetables or spices, as well as species utilized as seasoning crops, traditionally used medications, and ornamental plants (Fritsch and Friesen, 2002).

The main focus of investigators has been on existing natural plants that are recently cultivated in investigations, and most of Allium newly introduced species have been found in Iran. A number of recent species and subspecies achieved scientific qualification from some regions in Iran (Akhani 1999; Fritsch et al. 2001; Fritsch et al. 2002; Fritsch et al. 2006; Fritsch and Abbasi 2008;Fritsch and Maroofi 2010;Kamelin and Seisums 1996;Khassanov and Memariani 2006;Khassanov et al. 2006;Mashayekhi et al. 2005;Memariani et al. 2007;Neshati et al. 2009;Razyfard et al. 2011). A comparison was made among 170 species and subspecies, and their diverse types in Southwest Asia, mainly in Iran and Turkey, with over 120 species identified in Iran classified into seven subgenera and 30 sections (Friesen et al. 2006;Fritsch and Maroofi 2010;Fritsch and Abbasi 2013;Memariani et al. 2012). Allium genus is generic for the Irano-Turanian Phyto-geographic zone and represents a highly endemic rate (Matin 1992).

To increase the number of qualitative traits for tightly related species, molecular markers have greater capability than morphologic characters, most of which present only quantitatively diverse attributes (Harpke et al. 2013). Linne von Berg pioneered in a publication for the organization of the Allium genus by molecular markers (von Berg et al. 1996). Friesen then innovated the classification of Allium in publication according to molecular techniques (Friesen et al. 2006). Thereafter, Allium was affirmed monophyletically by the entire scientific reports (Choi et al. 2012;Li et al. 2010;Nguyen et al. 2008). Despite a huge body of research, perfect studies have not been conducted on the whole subgenera regarding the phylogenetic condition of their species (Friesen et al. 2006).

Currently, the DNA barcoding technique has been proven as an instructive and efficacious procedure for assessing plant phylogenies and presented successful applications in the molecular classification of Allium (Abdulina, 1999;von Berg et al. 1996). The basis of barcoding technique is on the alignment of short sequences of DNA markers of the nuclear and plastid genomes (Kress, 2017;Li et al. 2015). The analysis of phylogenetic associations and plant identification surveys have initially used variabilities of the nucleotide sequence in plastid DNA (cpDNA) at inter-specific (among family or genus) and intra-specific levels (into the species or varieties) (Tamura et al. 2004).

The matK is a chloroplast gene that encodes a locus in the intron of the trnk gene region. The gene encodes a maturase on the large single-copy part at the adjacency of the inverted repeat of plant species. The matK has vast substitution rates in comparison to other chloroplast genes, and its gene sequence lies among the lowest conserved plastid genes; hence, it has received effective uses in plant evolution and solving phylogenetic equations in a variety of taxonomic levels (Fuse and Tamura, 2000;Ito et al. 1999).

matK gene is very beneficial over other genes that contain the organelle genome genes. Firstly, this gene undergoes evolution nearly thrice as fast as the vastly applied plastid genes such as rbcL and atpB . matK gene is in the chloroplast genome and, in general, it has maternal inheritance. The gene is sensibly sizeable, with an extensive substitution rate, a large ratio of alterations at the first and the second codon positions, a low transition/transversion ratio, and the occurrence of mutationally conserved segments. It is also powerful and efficacious in species discrimination, and has excellent sequence recovery rate, a simple technique experimentally, a facile sequence alignment, and nonexistence of allelic polymorphisms or manifold paralogous copies against the nuclear DNA genome. It was demonstrated that the conversion at nucleic acid (DNA) and amino acid levels have even distribution throughout the whole gene. Apparently, the 5’ region of the matK gene contains a greater variation than the 3’ region in some monocotyledons and dicotyledons. To address family and even species-level associations, matK gene sequences (at both nucleic acids and amino acid sequence levels) has had successful applications for these specific attributes (Brochmann et al. 1998;Burgess et al. 2011;Hollingsworth et al. 2009;Koch et al. 2001;Lahaye et al. 2008;Steele and Vigalys, 1994;Tamura et al. 2004). To distinguish numerous plant species, the matK DNA gene has been utilized as a barcode gene alone or combined with other plant barcode sequences in current investigations (Bandara et al. 2013;De Mattia et al. 2011;Guo et al. 2011;Jing et al. 2011;Lahaye et al. 2008;Li et al. 2011;Pang et al. 2011;Seberg et al. 2012;Zhi‐Yuan et al. 2011).

Plant materials

Totally, 46 species in 11 sections representative of five subgenera of Allium were gathered from their natural habitat and relocated to the Iranian Biological Resource Center (IBRC), located in Karaj province of Iran during 2017-2018 (Table 1). The identification of the specimens was according to morphologic traits and diagnostic explanations of the species in related publications (Fritsch and Abbasi, 2013; Wendelbo, 1971). The genus was classified according to up-to-date infrageneric concepts presented previously (Friesen et al. 2006,Fritsch et al. 2010, andFritsch and Abbasi 2013).

Table 1 . Allium species analyzed using nucleotide sequences of matK DNA regions.

Code	Species	Subgenus	Section	Accession number
1	A. akaka sub akaka	Melanocrommyum	Acanthoprason	P1009858
2	A. akaka sub bozghushense	Melanocrommyum	Acanthoprason	P1011312
3	A. alamutense	Melanocrommyum	Acanthoprason	P1010457
4	A. ampeloprasum sub ampeloprasum	Allium	Allium	P1010913
5	A. ampeloprasum sub porrum	Allium	Allium	P1010908
6	A. asarence	Cepa	Cepa	P1004474
7	A. atroviolaceum	Allium	Allium	P1006775
8	A. austroiranicum	Melanocrommyum	Acanthoprason	P1009666
9	A. bisotunense	Melanocrommyum	Melanocrommyum	P1011153
10	A. breviscapum	Melanocrommyum	Acanthoprason	P1010598
11	A. breviscapum	Melanocrommyum	Acanthoprason	P1010598
12	A. cepa	Cepa	Cepa	P1010909
13	A. cf. assadi	Melanocrommyum	Megaloprason	P1011015
14	A. cf. cardiostemon	Melanocrommyum	Melanocrommyum	P1009998
15	A. cf. latifolium	Melanocrommyum	Acanthoprason	P1010022
16	A. chrysantherum	Melanocrommyum	Melanocrommyum	P1010382
17	A. derderianum	Melanocrommyum	Acanthoprason	P1009475
18	A. egorovae	Melanocrommyum	Acanthoprason	P1011306
19	A. fistulosum	Cepa	Cepa	P1010910
20	A. haemanthoides	Melanocrommyum	Acanthoprason	P1011119
21	A. iranicum	Allium	Allium	P1009733
22	A. iranshahrii	Melanocrommyum	Acanthoprason	P1011271
23	A. kazerouni	Melanocrommyum	Procerallium	P1010341
24	A. keusgenii	Melanocrommyum	Melanocrommyum	P1010423
25	A. koelzii	Melanocrommyum	Pseudoprason	P1010355
26	A. latifolium	Melanocrommyum	Acanthoprason	P1011289
27	A. mahneshanense	Melanocrommyum	Acanthoprason	P1010524
28	A. materculae	Melanocrommyum	Acanthoprason	P1009955
29	A. cf. minutiflorum	Melanocrommyum	Acanthoprason	P1010986
30	A. moderense	Melanocrommyum	Melanocrommyum	P1011021
31	A. mozaffarianii	Melanocrommyum	Melanocrommyum	P1011232
32	A. pesodoholandicum	Melanocrommyum	Procerallium	P1010545
33	A. psedoampeloprasum	Allium	Allium	P1009877
34	A. rubellum	Allium	Avulsea	P1009972
35	A. sabalanense	Melanocrommyum	Acanthoprason	P1009837
36	A. sativum	Allium	Allium	P1010653
37	A. scabriscapum	Reticulatobulbosa	Scabriscapa	P1009668
38	A. scabriscapum	Reticulatobulbosa	Scabriscapa	P1009668
39	A. stamineum	Allium	Codonoprasum	P1009964
40	A. stipitatum	Melanocrommyum	Procerallium	P1010429
41	A. sub akaka	Melanocrommyum	Acanthoprason	P1009965
42	A. tripedale	Nectaroscordum	Nectaroscordum	P1009676
43	A. ubipetrense	Melanocrommyum	Acanthoprason	P1010544
44	A. umbilicatum	Allium	Avulsea	P1009439
45	A. zagricum	Melanocrommyum	Acanthoprason	P1011079
46	Allium Sp.	Melanocrommyum	Acanthoprason	P1009928

DNA extraction, PCR Amplification and DNA Sequencing

The entire DNA was extracted from fresh leaf tissues of Allium species by the “IBRC plant extraction kit” (IBRC IND.) as recommended by the manufacturer’s protocol. The extracted DNA was assessed spectrophotometrically in terms of concentration and quality by determination of the absorbance at 260 nm and 280 nm. Different concentrations of template DNA and Mg and different annealing temperatures were used to optimize the PCR condition. Next, the PCR reaction materials were set in a final volume of 50 µL using Taq DNA Polymerase Master Mix RED (Amplicon), 1 ng of the primers, and 0.5 µL of genomic DNA.

The matK region was amplified using the primers 5’-CGA TCT ATT CAT TCA ATA TTT C- 3’ and 5’- TCT AGC ACA CGA AAG TCG AAG T -3’, matK 390F and matK 1326R primers for matK gene. DNA was amplified on an Eppendorf Master cycler gradient (Eppendorf Scientific, Germany) with the setting below: initial denaturation for 2 min at 92°C, followed by 35 cycles of denaturation (94°C, 1 min), annealing (55°C, 1 min), extension (72°C, 55 sec); and a final extension for 5 min at 72°C. After completion of the PCR reaction, 5 µl of PCR solution with 5 µL loading buffer was decanted into 1.5% agarose gel well holding TBE buffer. The PCR products were subjected to electrophoresis at 90 V for 85 min. Thereafter, the agarose gel was stained for 20 min in 0.50 mg/l of ethidium bromide and replicated fragments were visualized under UV light followed by gel document imaging. These PCR products were purified by the IBRC DNA purification kit (IBRC, Iran) based on the manufacturer’s instructions. A 10 kbp DNA ladder (Thermo Scientific, USA) was utilized as a molecular size standard (Fig. 1). PCR amplification was redone two times or sometimes more for each primer to ensure that the results were reproducible. The sequencing products were produced with the Sanger method by MWG Co. (Germany) that performed sequencing in both directions by the PCR primers.

Figure 1. Electrophoresis of purified PCR products of matK gene from Allium species

Editing, Sequence Alignment and Phylogenetic Reconstruction

Extraction of DNA sequences was performed from the chromatograms of the company by the use of Chromas v2, which were edited by the BioEdit program (Hall et al. 1999). The sequences were then put together with the CAP3 tool (Huang et al. 1999) and alignment of the homologous sequences was carried out using the EMBL-EBI CLUSTAL W tool (Edgar, 2004). The MEGA v6.0 software was used to analyze the multiple sequence alignment (MSA) file (Tamura et al. 2011). The genetic distance per loci among the accessions was estimated based on the number of base-pair replacement among the sequences. All the positions with missing data were eliminated by Kimura’s 2-parameter model. Moreover, the distance between matrices from the three loci by DNAsp was analyzed with Pearson’s correlation. The phylogenetic tree was developed by the Maximum Parsimony (MP) and Neighbor Joining (NJ) approach with a 1000 replicate bootstrap by the MEGA 6.0 software.

For estimating the resolution of DNA barcode, the percentage of produced monophyletic groups was determined using a bootstrap greater than 50% as a factor for defining the nodes, as recommended by Tripathi in DNAsp (Tripathi et al. 2013). The Tajima’s D and Fu’s Fs naturality tests were calculated by the DNAsp software. The amount of the dN/dS ratio was obtained numerically by the use of HIV databases. Haplotype network was scrutinized by the popART software (Leigh and Beryant, 2015).

The scientific progressive innovations in molecular science and sequencing approaches has empowered the recognition of organismal genomes. Besides, important data are provided by continuing variety of ongoing genome projects for multiple species concerning their classification, gene structure, and application scientifically. Here, nucleotide polymorphisms of the matK gene quality are applied for 46 species of Allium to recognize the levels and patterns of interspecific and distinctions.

According to our findings on the analyzed matK gene sequence in Allium species, thymine (38.5%) and guanine (13.9%) bases had the most remarkable and the lowest nucleotide rates (Table 2). In the nucleotide substitution recorded in the Allium species, substitution rates were detected significantly in pyrimidine, namely 26.22 for C-T and 10.68% for T-C conversions (Table 3). The above levels were less for purines for the G-A and A-G conversions (19.22% and 8.08%, respectively). These findings correspond to those of other investigators reporting consecutive cases of pyrimidine substitutions, which most probably result from cytosine methylation (Picoult et al. 1999).

Table 2 . Nucleotide abundances derived from the matK gene of Allium species.

G	C	T/U	A	Nucleotide
13.9	15.2	38.5	32.4	Frequency

Table 3 . Nucleotide substitution pattern estimation matrix of the matK gene in Allium species.

G	C	T	A	From/To
8.08	2.75	6.74	-	A
2.49	10.68	-	5.92	T
8.49	-	26.22	5.92	C
-	2.75	6.74	19.22	G

Each value represents the frequency of substitutions from one base (row) to another base (column). In this table, the percentage of transition mutations (purine-purine substitution, pyrimidine-pyrimidine substitution) and transversion mutations (purine-pyrimidine substitution and vice versa) are shown in bold and italics, respectively..

A total of 595 mutations was identified for genetic indicators of the matK gene in Allium species, which had different distributions all over the genome. Polymorphisms were detected in 243 sites, demonstrating the process of a positive selection of the matK gene sequence (Table 4). Searching for conserved DNA sections of the matK gene in the Allium species revealed a 0.26 CT region, a MWL of 50 bases, and a sequence conservation of 0.162 (Table 5). These conserved sections comprised a minor portion of the matK gene sequence, indicating that this site has discriminated differently and is susceptible to nucleotide alterations and mutations among various species, leading to variability among species.

Table 4 . Gene polymorphism of matK location in Allium species.

K	Eta	Pi	Hd	H	S	Population
80.112	595	0.160	0.9874	39	423	Allium

S: Number of polymorphic positions, H: number of haplotypes, Pi: nucleotide diversity, Eta: total number of mutations, K: number of nucleotide differences between populations or species (nucleotide divergence).

Table 5 . Conserved DNA regions of thematK gene in Allium species.

CT	MWL	C	Population
0.26	50	0.162	Allium

C: Sequence conservation, MWL: Minimum conservation Length, CT: Conservation threshold.

An estimated value of 1.16 was obtained for the dN/dS proportion in Allium species (Table 6), suggesting the positive selection of the matK gene in Allium species in the course of evolution. This kind of evolution has led to new species and stabilized better refinement of their efficacy during the evolution, which arises from the conversion of non-coding gene sites to the gene coding sites. The same as Tajima's D and Fu's Fs, neutrality tests were obtained to determine deviations from the null theory on the neutral evolution and identify the impacts of natural selection on these genes in Allium species. Significantly negative estimates of D (-1.50) and Fu's Fs (-1.19) are made by groups of people under the influence of recently advanced or highly developed effective population size or the directional selection. Positively estimated values of D and Fs indicate the influences of the genetic drift, genetic dilemma, or a balancing effect over the evolutionary history of the population. The findings of the present research proved negative estimates of D and Fs (Table 7). The negative and positive results represent a significantly slight dissimilarity between polymorphisms with regard to their frequency, respectively. According to the outcomes of both neutrality tests, the continuing development of the Allium species has affected the planet or directional selection has influenced this gene throughout evolution. Altogether, Fu's Fs and Tajima's D tests have been proven to have higher effectiveness for small size and larger estimated populations.

Table 6 . Identification of the matK gene natural selection process in Allium species.

Numerical value	Parameter
2.3254	dN
1.9970	dS
1.1644	dN/dS†

^†The numerical value of dn / ds represents the natural selection process.

Table 7 . Results obtained from an evaluation of the natural evolution of the matK gene in Allium species.

Fu's Fs	Tajima's D	Gene
-1.19	-1.50	matK

It is of paramount importance to determine haplotype groups (by the popART software) for determination of the geographic regions of the examined breeds in comparison to other breeds. The entire 46 samples grouped in haplotype bunch B is the largest in various species around the world. Based on our observations, the haplogroup A is usually present in all continents, and the haplogroup B may similarly have originated from Asia (Fig. 3) (Ghanbari et al. 2018).

Figure 3. Interconnectivity grid of the varieties of Allium species studied using the matK gene (popART software)

The NJ tree was drawn by the Kimura distance (determination of distances aimed at building NJ tree using the MEGA 6.0 with the program defaults). The nucleotide sequences of the matK site presented no significantly different topology from the MP tree (MP tree was produced using the MEGA 6.0 by the program defaults), which differed in their branch-length (Kimura, 1980;Nei and Kumar, 2000).

In the current investigation, the matK gene sequence of Amaryllis belladonna was utilized for the outside group. Phylogenetic assays that depended on the nucleotide sequences of matK could mostly differentiate subgenera and sections in the genus Allium ; though, few accessions of species were located outside of the section. Three discrete clades were produced by the NJ dendrogram. The first clade consisted of species A. austroiranicum and A. ampeloprasum. The second clade comprised the species A. iranshahrii , A. bisotunense , A. cf assadi , and A. rubellum , and the rest of species lied in the third clade.

A. egorovae presented middle sites between groups of the first and the second evolutionary lines in the tree topology. The produced matK phylogenetic tree endorsed the theory of Allium as a monophyletically originated genus. Some species analyzed phylogenetically for the matK site revealed that accessions were placed in distant clades. A. ampeloprasum lied in a clade near A. umbilicatum and another accession of A. ampeloprasum had a place in a distant clade with A. austroiranicum . The majority of the species were also distinguished by matK , and few species with close relations were not distinguishable. A. zagricum and A. alamutense lied in a similar clade without any distinct dissimilarities.

In a previous study, the matK gene was proposed for resolving phylogenetic associations in plant species (Abugalieva et al. 2017;İpek et al. 2014,Ito et al. 1999;Kim et al. 2018;Son et al. 2010). The genes utilized for DNA barcoding in plant populations include matK , rpoC1 , rpoB , trnH-PsbA , rbcL , atpF-atpH , psbK-psbI , and their combinations, of which, matK and rbcL were adopted as 2-locus DNA barcode by the CBOL Group (Hollingsworth et al. 2009). Polymorphic barcode nucleotide sequences are desirable at interspecific or upper classification levels, but not at the intraspecific level. Accordingly, comparison of the barcode nucleotide sequence can identify an unidentified plant accession as a species (Hebert et al. 2003;Kress and Erikson, 2008;Pang et al. 2011;Stoeckle, 2003). Additionally, the DNA fragment used as barcoding should have high recoverability, high proportion in species identification, and proper affordability (Burgess et al. 2011).

Here, the efficacy of the matK gene was examined as a DNA barcode for discriminating Allium species. Our findings indicated that matK sites could be easily amplified by PCR. DNA sections were mostly discrete parts in Allium . Nevertheless, some species were not distinguishable from one another by the matK sites (Abugalieva et al. 2017;Ince et al. 2005;İpek et al. 2014;Li et al. 2011;Son et al. 2010).

The length of matK nucleotide sequences varied between 540 bp in A. porrum and A. austroiranicum and 1023 bp in Allium sp. Genetic associations among Allium species were examined through the polymorphisms within the nucleotide sequences of trnH-psbA , ITS and matK in a recent report (İpek et al. 2014;Son et al. 2010). Over one haplotype in both organellar and nuclear genomes were present in a single plant of Allium species, making phylogeny analysis and barcoding implausible applying the nucleotide arrays of these DNA sites. At present, 39 haplotypes were noticed in Allium species for matK in the plastid genome.

Phylogenetic associations among Allium species on the basis of matK analyses are in line with earlier research by the use of the matK site (Abugalieva et al. 2017;Gurushidze et al. 2007;İpek et al. 2014;Li et al. 2010). A. asarence and A. cepa lied jointly in a group with no polymorphisms in some investigations (Li et al. 2010;Nguyen et al. 2008). Likewise, A. asarence and A. cepa put together in a cluster by analyzing matK in the present research. Moreover, an accession of A. ampeloprasum was tightly grouped with A. umbilicatum and A. atroviolaceum , which is almost similar to those of Nguyen et al. and Hirschegger et al. On the other side, an accession of A. ampeloprasum (var. porrum) lied in close grouping with A. austroiranicum.A. Akaka was placed tightly in a group with A. breviscapum and A. sub akaka was assigned to a group close to A. alamutense and A. zagricum , which have a closer relation to a recently published report (Akhavan et al. 2015;Li et al. 2010;Nguyen et al. 2008). Similar to a latest investigation, A. pseudoampeloprasum was clustered tightly to A. atroviolaceum , A. iranicum, and A. sativum in our research ((Hirschegger et al. 2010;Veiskarami et al. 2019). A. umbilicatum lied in a group with A. ampeloprasum and A. atroviolaceum , which corresponds to a latest research (Friesen et al. 2006;Li et al. 2010). A. fistulosum was in group close to A. cepa and A. asarence as reported previously (Veiskarami et al. 2019;Li et al. 2010;Friesen et al. 2006). A similar finding was reported in an earlier study (Akhavan et al. 2015;Friesen et al. 2006;Fritsch and Abbasi, 2013;Fritsch et al. 2010;Gurushidze et al. 2010;Li et al. 2010;Sýkorová et al. 2006). Two accessions of A. scabriscapum were positioned in a similar clade and clustered closely with A. asarence , A. cepa , and A. fistulosum by analyzing matK , which is in agreement with a recently studied case (Friesen et al. 2006;Veiskarami et al. 2019). As detected in a recently published study, A. sub Akaka , A. zagricum, and A. alamutense were put together with no distance (Akhavan et al. 2015;Fritsch et al. 2010; Gurushidze et al. 2008;Li et al. 2010).

These observations indicated a close relation between Allium species in subgenera Allium and Reticulatobulbosa. Likewise, it was recently demonstrated that species in these subgenera were closely related phylogenetically (Friesen et al. 2006;Li et al. 2010). All accessions of Allium species in subgenus Allium were grouped in a similar clade. Yet, A. rubellum and A. stamineum from subgenus Allium were placed in the clade of subgenus Melanocrommyum and Reticolatobulbosa with matK examinations. This finding suggests that mixtures are present in these accessions. Accessions relating to the subgenus Melanocrommyum were put together in a clade and endorsed by a 100% bootstrap value. Similarly, the present outcomes according to the matK site revealed that these species had close phylogenetic relations. To conclude on the basis of matK examination, phylogenetic associations among Allium species scrutinized here corresponded to those reported previously (Li et al. 2010;Veiskarami et al. 2019).

Here, A. mahneshanense was grouped with A. zagricum whereas it was clustered with A. cf. minutiflorum elsewhere. A. cf. minutiflorum was gathered with A. moderense while it was assigned to another Allium spp. elsewhere. In the current research, A. iranshahrii , A . bisotunense , and A. cf. assadi were put together tightly whereas A. bisotunense was clustered with A. keusgenii and A. iranshahrii was assigned closely to A. haemanthoides in other investigations. (Akhavan et al. 2015; Gurushidze et al. 2008;Li et al. 2010). It is proposed to utilize matK as an additional means for analyzing Allium phylogenetically as using the matK site had more simplicity for recouping and more affordability than other sites. The present information indicated that the sampled Allium species were related to the subgenera in the second and third evolutionary lines (Friesen et al. 2006;Fritsch and Abbasi 2013). The dendrogram in the matK examination recommends that species in the subgenus Melanocrommyum developed earlier than in the subgenus Allium , and these two subgenera possess a shared genetic node (Fig. 2).

Figure 2. Neighbor-joining phylogenetic tree analysis of 46 Allium species based on the nucleotide sequences of a matK region

As in the phylogenetic tree drawn inFriesen et al. (2006) in which subclades Polyprason , Reticulatobulbosa , and Cepa formed three sister subclades, three subclades were detected herein. The first subclade was species from the subgenera Melanocrommyum , whereas the succeeding one contained species of Allium and cepa . Our presented phylogeny of matK suggests that five subgroups can be distinguished within subgenus Melanocrommyum . A. stipitatum , A. kazerouni , and A. pseudoholandicum are the representatives of the first group (section procelarium). A. chrysantherum , A . mozaffarianii , A. cf. cardiostemon, A. keusgenii, A. moderense , and A. bisotunense (section Melanocrommyum) comprise the second group, A. koelzii (section pseudoprason) belongs to the third group, A. cf. assadi (section megaloprason) is related to the fourth group, and the rest of species from section Acanthoprason form the fifth group.

A research confirmed the Allium genus originated monophyletically, which was proved in some other literature (Friesen et al. 2006;Li et al. 2010;Nguyen et al. 2008). Additionally, a thorough research compared Melanocrommyum species and mainly focused on the morphological and molecular genetic explanation of species cultivated in Iran. In spite of the fact that Fritsch and Abbasi (Fritsch and Abbasi, 2013) studied fundamentally the phylogenetic taxonomy of the Allium genus, lots of ill-known Allium taxa remain accessible in all around the country. To evaluate the phylogeny of 46 endemics, scarce Allium species with economic importance from Iran, the matK DNA barcoding marker was employed in this research.