Calicioids of Alberta, Canada, with descriptions of three new genera and 13 new Sphinctrinaceae species to science

We conducted a floristic revision of calicioid lichens and allied fungi from Alberta, Canada, including the descrioption of 13 species new to science, seven of which were assigned to three newly described genera (Albocalicium candidum, Brevicalicium roseum, Paracalicium betulae, P. caraganae, P. chamaedaphnes, P. piceae, and P. recedens) and six to two existing genera (Chaenothecopsis abscondita, C. caelumsaltator, C. calicii-viridis, C. epifurfuracea, C. yukonensis, and Phaeocalicium alnophilum). Seventy-three new sequences from 57 specimens across 34 species were generated (43 ITS, 9 LSU, and 21 mtSSU), including sequences for all newly described species and the first published sequences for Chaenotheca selvae, Chaenothecopsis ochroleuca, C. penningtonensis, C. parasitaster, and Phaeocalicium flabelliforme. We provided phylogenies for the Caliciaceae, Coniocybaceae, and Sphinctrinaceae, using all available public sequences. Novel clades and traits in Chaenotheca phaeocephala and Stenocybe pullatula are explored. Here we deposit the individual alignments for each locus and family, as well as the concatenated matrices (minus excluded regions) used to generate our phylogenies.

Methods
Molecular methods: To test species hypotheses, we generated new sequences of three loci from material in Alberta, Northwest Territories, and Yukon Territory: the fungal genetic barcode, the internal transcribed spacer (ITS, nuclear ribosomal DNA consisting of the internal transcribed spacer regions 1 and 2, the embedded 5.8S region, and small sections of the flanking large and small ribosomal subunits, LSU and SSU), a longer segment of the nuclear ribosomal large subunit (LSU), and the mitochondrial small subunit (mtSSU). Sanger sequencing was carried out in the Canadian Museum of Nature’s Laboratory of Molecular Biodiversity (RDB), the Spribille Laboratory, and the Molecular Biology Service Unit at the University of Alberta (EP). We used previously unpublished sequences generated by the Canadian Centre for DNA Barcoding (BOLD, https://boldsystems.org/) as part of the Arctic Probe Project, as well as most NCBI sequences of calicioids from the three target families.

DNA extraction, PCR amplification, and sequencing: Canadian Museum of Nature: Genomic DNA was extracted from dried material following a silica column purification protocol similar to commercially available DNA extraction kits (modified from Alexander et al. 2007). DNA extraction success was assessed via gel electrophoresis on 1.25% agarose gels stained with ethidium bromide. After extractions, the three gene loci were amplified with the following primers: ITS with ITS1, ITS2, ITS3 and ITS4 (White et al. 1990); mtSSU with mrSSU1, mrSSU2, mrSSU2R and mrSSU3R (Zoller et al. 1999), and LSU with LIC24R (Miadlikowska & Lutzoni 2000), LR0R, LR3, LR3R, LR5, LR6 and LR7 (Vilgalys & Hester 1990).

DNA was amplified using polymerase chain reaction (PCR) in a 15 μL volume with 9.05 μL of DNA-grade H20, 3 μL of 5x reaction buffer, 0.3 μL of 10 mM dNTP, 0.75 μL of 10 μM each primer, 0.3 U of Q5 DNA Polymerase (New England BioLabs Inc.), and 1 μL of DNA template. Some LSU and mtSSU amplifications were carried out using DreamTaq DNA Polymerase (ThermoFisher Scientific) in a 15 μl volume with 11.3 μL of DNA-grade H20, 1.5 μL of 10x reaction buffer, 0.3 μL of 10 mM dNTP, 0.375 μL of 10 μM each primer, 0.75 U of polymerase, and 1 μL of DNA template. DNA template was doubled for difficult samples. For Q5 amplifications, an initial denaturation of 98°C for 30 sec was followed by 34 cycles of 98°C for 10 sec, 56°C for 30 sec, 72 °C for 30 sec and a final extension step of 72°C for 5 min. For DreamTaq amplifications, an initial denaturation of 95°C for 3 min was followed by 35 cycles of 95°C for 30 sec, 55°C for 30 sec, 72 °C for 90 sec and a final extension step of 72°C for 10 min. Amplification success was assessed via gel electrophoresis in 1.25% agarose gels stained with ethidium bromide. Sequencing reactions were performed in 10 μL reactions containing 6.2 μL of DNA-grade H20, 1.8 μL of 5x reaction buffer, 0.5 μL of primer, 0.5 μL of BigDye Terminator v3.1 Ready Reaction Mix (ThermoFisher Scientific), and 1 μL of diluted PCR products. An initial denaturation of 95°C for 3 min was followed by 30 cycles of 96°C for 30 sec and 50°C for 20 sec followed by a final step at 60°C for 4 min. Reaction products were purified via an EDTA-NaOH-ethanol precipitation protocol recommended by the sequencing kit manufacturer. Purified DNA pellets were resuspended in HIDI formamide, denatured at 95°C for 5 min, cooled for 2 min, and sequenced via automated capillary electrophoresis on an Applied Biosystems 3500xL Genetic Analyzer (ThermoFisher Scientific).

DNA extraction, PCR amplification and sequencing: University of Alberta, Genomic DNA was extracted from dried material using the QIAamp DNA Investigator kit (Qiagen, Hilden, Germany) following the manufacturer’s protocols for isolation of total DNA from tissue, except samples were first lysed with a mechanical bead beater for 30 sec and fragments were incubated in lysis buffer for 8 hours at 56℃. After extractions, samples were quantitated with a Nanodrop spectrophotometer (Implen NP80), and the three gene loci were amplified with the following primer sets: ITS with ITS1F (Gardes & Bruns 1993) and ITS4 (White et al. 1990), mtSSU with mrSSU1 and mrSSU3R (Zoller et al. 1999), and LSU with LR7 and LR0R (Vilgalys & Hester 1990).

DNA was amplified using PCR in 22 μL reactions for each gene locus of interest. Amplification of each gene region was performed as follows: ITS amplification used an initial denaturation at 95℃ for 5 min, and then 35 cycles of 95℃ for 30 sec, annealing at 57℃ for 30 sec and then extension at 72℃ for 30 sec, followed by a final extension at 72℃ for 7 min and holding at 4℃. LSU amplification used an initial denaturation at 95℃ for 5 min and then 35 cycles of 95℃ for 1 min, annealing at 56℃ for 1 min, 72℃ for 1 min 30 sec and then a final extension at 72℃ for 7 min and holding at 4℃. Finally, for mtSSU amplification an initial denaturation at 95℃ for 5 min and then 35 cycles of denaturation at 95℃ for 1 min, annealing at 54℃ for 1 min, extension at 72℃ for 1 min and then a final extension at 72℃ for 7 min and holding at 4℃. Once visualized using gel electrophoresis using 1% agarose to ensure PCR success and cleaned using a 4 μl ExoSAP reaction to remove unused primers and nucleotides, amplicons were sent for Sanger sequencing (ABI 3730, Thermo Fisher) at the University of Alberta Molecular Biology Service Unit.

Sequence screening and selection: We initially included almost all unique published sequences for the three families in NCBI. We screened both new and published sequences with ‘megaBLAST’ searches against the NCBI nucleotide database to identify sequences that may represent non-target organisms and to identify the closest published relatives (NCBI Resource Coordinators 2018). Because of gene tree conflict and difficulty with de novo sequencing, all newly generated and published sequences in Sphinctrinaceae were analyzed in T-BAS, a tree-based alignment selector toolkit (https://tbas.cifr.ncsu.edu/tbas2_3/pages/tbas.php, Carbone et al. 2017, 2019; Miller et al. 2015). We checked the placement of both individual loci and concatenated sequences within Pezizomycotina using the ‘place unknowns’ and environmental phylogenetic placement analyses. While we treat the name Sphinctrinaceae as synonymous with Mycocaliciaceae following Jaklitsch et al. (2016) and Ertz et al. (2023), the T-BAS reference tree retains Mycocaliciaceae as distinct from Sphinctrinaceae (Carbone et al. 2017). Subsequently, we excluded some published sequences because they i) blasted to non-calicioid sequences, ii) had low percent identities with any other sequences in GenBank (<88%), iii) T-BAS placed them in a family outside of the class Eurotiomycetes, and/or iv) they did not cluster with any of our putative new species using a local blast. For ITS, we excluded: Chaenothecopsis orientalis accession AY795863, C. rubescens OQ717807, C pusilla OQ717806, AF243132, AY795866 (placement variable, remote from Sphinctrinaceae); Mycocalicium victoriae AF243135, AJ312123, AY128702, AY128701, and M. sp. MN206996, MT558584, AJ972853 (97–98% percent identity to M. victoriae, placed in class Dothideomycetes, family Teratosphaeriaceae with high likelihood); Sphinctrina intermedia KJ865747 (placed in class Lecanoromycetes, clustered with Circinaria and Aspicilia, and the text in Tibell et al. (2014) states this sequence was an attempted sequencing of Phaeocalicium triseptatum; Tibell confirmed our interpretation, pers. com. Nov. 2024); Chaenotheca stemonea AF408683 and C. brunneola OQ843252. For LSU we excluded Chaenothecopsis hunanensis JX122784, C. proliferatus JX122783, and Phaeocalicium praecedens KC590486 (placed in Dothideomycetes). Finally, we concluded that accession KJ871615 (labeled in NCBI as Phaeocalicium triseptatum) represents Sphinctrina intermedia following the text in Tibell et al. 2014 and supported by our phylogenetic analyses (Tibell confirmed our interpretation, pers. com. Nov. 2024). We also excluded nine de novo sequences for similar reasons (5 LSU, 3 ITS, 1 mtSSU).

We report similarity metrics with accessioned sequences for newly generated loci when material is concordant morphologically with described species. We constructed de novo phylogenetic trees for Caliciaceae, Coniocybaceae and Sphinctrinaceae to confirm phylogenetic relationships in both described and putative new species. We used two species of Heterodermia as outgroups for Caliciaceae, following Prieto & Wedin (2017). For the Coniocybaceae phylogeny we adopted outgroups from Suija et al. (2023) and Tibell et al. (2019). For Sphinctrinaceae, we adopted outgroups from orders outside of Mycocaliciales following Ertz et al. (2023; adopted from Beimforde et al. 2023, 2017; Prieto et al. 2013; Thiyagaraja et al. 2022; Tibell & Vinuesa 2005; Tuovila et al. 2013; Vinuesa et al. 2001). Supplementary Table S2 provides information on the final 475 sequences used here. Across the three analyses, we included published sequences from: Aguirre-Hudson et al. 2007; Ariyawansa et al. 2015; Beimforde et al. 2023; Crous et al. 2013; De Leo et al. 2003; Ertz et al. 2023; Etayo et al. 2023; Hanani et al. 2022; Hofmeister et al. 2022; Houbraken et al. 2011; James et al. 2006; Kauff et al. 2018; Li et al. 2023; Lutzoni et al. 2001; Malíček 2022; Marthinsen et al. 2019; Mark et al. 2016; Masumoto et al. 2019; McMullin et al. 2024; Messuti et al. 2012; Gaya et al. 2012, 2014; Ohmura et al. 2022; Pavlov et al. 2023; Povilaitienė et al. 2022; Prieto & Wedin 2017; Prieto et al. 2013, 2021; Prieto 2020; Pykälä et al. 2019; Réblová et al. 2017; Rikkinen et al. 2014; Samson et al. 2009; Schoch et al. 2014; Schmull et al. 2011; Wiklund & Wedin 2003; Selva et al. 2023b; Sert et al. 2007; Spatafora et al. 2006; Spribille et al. 2020; Sterflinger & Prillinger 2001; Suija et al. 2016, 2023; Telfer et al. 2015; Temu et al. 2019, 2024; Thiyagaraja et al. 2022; Tibell 2001a, 2001b, 2002, 2003, 2006, 2007; Tibell & Beck 2001; Tibell & Knutsson 2016; Tibell & Koffman 2002; Tibell & Vinuesa 2005; Tibell et al. 2014, 2019; Tuovila et al. 2011, 2013, 2014; Vinuesa et al. 2001; Vondrák et al. 2022, 2023; Vu et al. 2019; Wang et al. 2005; Wedin et al. 2002, 2005; Weerakoon et al. 2012; Williams & Tibell 2008; Yahr 2015; Yang et al. 2022.

Phylogenetic analyses: For each family and locus, sequences were aligned in MAFFT (version 7.49, Katoh et al. 2019, 2002; Katoh & Standley 2013) via Mesquite (version 3.7, Maddison & Maddison 1997–2021) using the G-INS-i method. Alignments were vetted visually and adjusted manually using the Mesquite ‘Highlight Apparently Slightly Misaligned’ option. We used ITSx 1.1 (Bengtsson-Palme et al. 2013) on the DeCifr platform to split sequences into ITS, small subunit, and large subunit files to aid in sequence and alignment vetting. Ambiguously-aligned regions were identified using GBLOCKS (Castresana 2000) within Mesquite, using less stringent criteria (minimum length of a block was set to 2, all characters were allowed gaps, the minimum number of sequences for a flank position was set to 51% of taxa with non-gaps at that position) and excluded from further analyses. Terminal ends and introns were also excluded. Loci were concatenated in Mesquite without the excluded regions. Original (with excluded regions) and final alignments are deposited in Dryad, and sequence voucher data are provided in Supplementary Table S2.

本研究对加拿大阿尔伯塔省的杯点衣类地衣（calicioid lichens）及其近缘真菌开展了区系修订工作，共计记述13个科学新种：其中7个隶属于3个新建立的属（*Albocalicium candidum*、*Brevicalicium roseum*、*Paracalicium betulae*、*P. caraganae*、*P. chamaedaphnes*、*P. piceae* 及 *P. recedens*），剩余6个隶属于2个已有属（*Chaenothecopsis abscondita*、*C. caelumsaltator*、*C. calicii-viridis*、*C. epifurfuracea*、*C. yukonensis* 及 *Phaeocalicium alnophilum*）。本研究从34个物种的57份标本中获得了73条新序列（其中ITS序列43条、LSU序列9条、mtSSU序列21条），涵盖所有新记述物种的序列，同时还首次发布了*Chaenotheca selvae*、*Chaenothecopsis ochroleuca*、*C. penningtonensis*、*C. parasitaster* 及 *Phaeocalicium flabelliforme* 的序列数据。本研究利用所有公开可用的序列数据，构建了杯点衣科（Caliciaceae）、筒壳菌科（Coniocybaceae）及Sphinctrinaceae的系统发育树。本研究还探讨了*Chaenotheca phaeocephala* 与 *Stenocybe pullatula* 中的新演化支与性状特征。本文存档了各基因位点、各科的单独比对文件，以及用于构建系统发育树的拼接矩阵（已剔除排除区域）。

研究方法

分子生物学方法：为验证物种假说，本研究从采自阿尔伯塔省、西北地区及育空地区的标本中扩增得到3个基因位点的新序列：真菌DNA条形码——内部转录间隔区（ITS，核核糖体DNA，包含内转录间隔区1、2，嵌入的5.8S区域，以及侧翼的大、小核糖体亚基LSU与SSU的小段序列）、核核糖体大亚基的更长片段（LSU），以及线粒体小亚基（mtSSU）。桑格测序（Sanger sequencing）分别在加拿大自然博物馆分子生物多样性实验室（RDB）、Spribille实验室以及阿尔伯塔大学分子生物学服务中心（EP）完成。本研究使用了加拿大DNA条形码中心（BOLD，https://boldsystems.org/）作为北极探测项目（Arctic Probe Project）一部分所产生的未公开序列，以及NCBI数据库中上述三个目标科的大部分杯点衣类真菌序列数据。

DNA提取、PCR扩增与测序：加拿大自然博物馆：采用基于硅胶柱纯化的方案（改编自Alexander等2007年研究），从干燥标本中提取基因组DNA。通过1.25%琼脂糖凝胶电泳结合溴化乙锭染色，检测DNA提取成功率。提取完成后，使用以下引物扩增三个基因位点：ITS区使用ITS1、ITS2、ITS3与ITS4引物（White等1990）；mtSSU区使用mrSSU1、mrSSU2、mrSSU2R与mrSSU3R引物（Zoller等1999）；LSU区使用LIC24R（Miadlikowska & Lutzoni 2000）、LR0R、LR3、LR3R、LR5、LR6与LR7引物（Vilgalys & Hester 1990）。

采用聚合酶链式反应（PCR）扩增DNA，Q5 DNA聚合酶（New England BioLabs公司）体系总体积为15 μL，包含9.05 μL DNA级纯水、3 μL 5×反应缓冲液、0.3 μL 10 mM dNTP混合液、每条引物各0.75 μL 10 μM溶液、0.3 U Q5 DNA聚合酶以及1 μL DNA模板。部分LSU与mtSSU区扩增使用DreamTaq DNA聚合酶（ThermoFisher Scientific），体系总体积15 μL，包含11.3 μL DNA级纯水、1.5 μL 10×反应缓冲液、0.3 μL 10 mM dNTP混合液、每条引物各0.375 μL 10 μM溶液、0.75 U DreamTaq DNA聚合酶以及1 μL DNA模板。对于扩增困难的样本，DNA模板用量加倍。Q5聚合酶扩增程序为：98℃预变性30秒，随后34个循环：98℃变性10秒、56℃退火30秒、72℃延伸30秒，最后72℃终延伸5分钟。DreamTaq聚合酶扩增程序为：95℃预变性3分钟，随后35个循环：95℃变性30秒、55℃退火30秒、72℃延伸90秒，最后72℃终延伸10分钟。扩增成功与否通过1.25%琼脂糖凝胶电泳结合溴化乙锭染色进行检测。

测序反应体系总体积为10 μL，包含6.2 μL DNA级纯水、1.8 μL 5×反应缓冲液、0.5 μL引物、0.5 μL BigDye Terminator v3.1预混液（ThermoFisher Scientific）以及1 μL稀释后的PCR产物。测序程序为：95℃预变性3分钟，随后30个循环：96℃变性30秒、50℃退火20秒，最后60℃保温4分钟。反应产物按照测序试剂盒制造商推荐的EDTA-NaOH-乙醇沉淀法进行纯化。纯化后的DNA沉淀用HIDI甲酰胺重悬，95℃变性5分钟后冷却2分钟，随后在Applied Biosystems 3500xL基因分析仪（ThermoFisher Scientific）上通过自动化毛细管电泳完成测序。

DNA提取、PCR扩增与测序：阿尔伯塔大学：使用QIAamp DNA Investigator试剂盒（Qiagen，德国希尔登），按照制造商提供的组织总DNA提取规程，从干燥标本中提取基因组DNA；实验中先通过机械珠磨破碎样本30秒，再将裂解片段置于裂解缓冲液中56℃孵育8小时。提取完成后，使用Nanodrop分光光度计（Implen NP80）对样本进行定量，随后使用以下引物组合扩增三个基因位点：ITS区使用ITS1F（Gardes & Bruns 1993）与ITS4（White等1990）引物，mtSSU区使用mrSSU1与mrSSU3R（Zoller等1999）引物，LSU区使用LR7与LR0R（Vilgalys & Hester 1990）引物。

每个目标基因位点的PCR扩增体系总体积为22 μL。各基因区域的扩增程序如下：ITS区扩增程序为：95℃预变性5分钟，随后35个循环：95℃变性30秒、57℃退火30秒、72℃延伸30秒，最后72℃终延伸7分钟并于4℃保温。LSU区扩增程序为：95℃预变性5分钟，随后35个循环：95℃变性1分钟、56℃退火1分钟、72℃延伸1分30秒，最后72℃终延伸7分钟并于4℃保温。mtSSU区扩增程序为：95℃预变性5分钟，随后35个循环：95℃变性1分钟、54℃退火1分钟、72℃延伸1分钟，最后72℃终延伸7分钟并于4℃保温。扩增产物通过1%琼脂糖凝胶电泳可视化以确认扩增成功，随后使用4 μL ExoSAP反应体系纯化以去除未使用的引物与核苷酸，最后将扩增子送至阿尔伯塔大学分子生物学服务中心，使用ABI 3730测序仪（Thermo Fisher）完成桑格测序。

序列筛选与选择：本研究最初纳入了NCBI数据库中上述三个科的几乎所有唯一已发表序列。通过对NCBI核苷酸数据库进行megaBLAST搜索，对新获得序列与已发表序列进行筛选，以识别非目标生物序列并确定最相近的已发表亲缘序列（NCBI Resource Coordinators 2018）。鉴于基因树冲突与从头测序的难度，本研究使用T-BAS（一款基于树的比对选择工具包，https://tbas.cifr.ncsu.edu/tbas2_3/pages/tbas.php，Carbone等2017、2019；Miller等2015）对Sphinctrinaceae的所有新生成序列与已发表序列进行分析。本研究通过‘未知序列定位’与环境系统发育定位分析，验证了单个基因位点序列与拼接序列在盘菌亚门（Pezizomycotina）中的分类位置。尽管按照Jaklitsch等（2016）与Ertz等（2023）的研究，我们将Sphinctrinaceae视为Mycocaliciaceae的同物异名，但T-BAS参考树仍将Mycocaliciaceae与Sphinctrinaceae视为两个独立的科（Carbone等2017）。

随后，本研究剔除了部分已发表序列，剔除标准包括：i）BLAST比对结果为非杯点衣类序列；ii）与GenBank中其他序列的相似性百分比较低（<88%）；iii）T-BAS分析将其定位在散囊菌纲（Eurotiomycetes）以外的科；以及/或iv）本地BLAST分析显示其未与本研究的推定新种聚为一支。针对ITS序列，本研究剔除了以下序列：登录号为AY795863的*Chaenothecopsis orientalis*、OQ717807的*C. rubescens*、OQ717806的*C. pusilla*、AF243132、AY795866（分类位置不稳定，与Sphinctrinaceae科亲缘关系较远）；AF243135、AJ312123、AY128702、AY128701的*Mycocalicium victoriae*，以及MN206996、MT558584、AJ972853的*M. sp.*（与*M. victoriae*的相似性为97%~98%，极大概率被定位在座囊菌纲（Dothideomycetes）的Teratosphaeriaceae科）；KJ865747的*Sphinctrina intermedia*（被定位在茶渍纲（Lecanoromycetes），与*Circinaria*及*Aspicilia*聚为一支，且Tibell等（2014）的文献中提及该序列为*Phaeocalicium triseptatum*的尝试测序结果；Tibell于2024年11月的个人通信中证实了本研究的判断）；AF408683的*Chaenotheca stemonea*与OQ843252的*C. brunneola*。针对LSU序列，本研究剔除了JX122784的*Chaenothecopsis hunanensis*、JX122783的*C. proliferatus*以及KC590486的*Phaeocalicium praecedens*（均被定位在座囊菌纲）。最后，本研究判定登录号KJ871615（NCBI中标注为*Phaeocalicium triseptatum*）实为*Sphinctrina intermedia*，该结论符合Tibell等（2014）的文献内容，且得到本研究系统发育分析的支持（Tibell于2024年11月的个人通信中证实了本研究的判断）。本研究还基于类似原因剔除了9条从头测序序列（其中LSU序列5条、ITS序列3条、mtSSU序列1条）。

当标本形态与已记述物种一致时，本研究报告了新生成基因位点与已登录序列的相似性指标。本研究为杯点衣科、筒壳菌科及Sphinctrinaceae构建了从头系统发育树，以验证已记述物种与推定新种的系统发育关系。参照Prieto & Wedin（2017）的研究，本研究选用2个*Heterodermia*物种作为杯点衣科的外类群。针对筒壳菌科的系统发育分析，本研究采用Suija等（2023）与Tibell等（2019）所使用的外类群。针对Sphinctrinaceae，本研究参照Ertz等（2023）的研究，选用Mycocaliciales以外的目类群作为外类群（数据源自Beimforde等2023、2017；Prieto等2013；Thiyagaraja等2022；Tibell & Vinuesa 2005；Tuovila等2013；Vinuesa等2001）。补充表S2提供了本研究最终使用的475条序列的相关信息。在三项分析中，本研究纳入的已发表序列源自以下文献：Aguirre-Hudson等2007；Ariyawansa等2015；Beimforde等2023；Crous等2013；De Leo等2003；Ertz等2023；Etayo等2023；Hanani等2022；Hofmeister等2022；Houbraken等2011；James等2006；Kauff等2018；Li等2023；Lutzoni等2001；Malíček 2022；Marthinsen等2019；Mark等2016；Masumoto等2019；McMullin等2024；Messuti等2012；Gaya等2012、2014；Ohmura等2022；Pavlov等2023；Povilaitienė等2022；Prieto & Wedin 2017；Prieto等2013、2021；Prieto 2020；Pykälä等2019；Réblová等2017；Rikkinen等2014；Samson等2009；Schoch等2014；Schmull等2011；Wiklund & Wedin 2003；Selva等2023b；Sert等2007；Spatafora等2006；Spribille等2020；Sterflinger & Prillinger 2001；Suija等2016、2023；Telfer等2015；Temu等2019、2024；Thiyagaraja等2022；Tibell 2001a、2001b、2002、2003、2006、2007；Tibell & Beck 2001；Tibell & Knutsson 2016；Tibell & Koffman 2002；Tibell & Vinuesa 2005；Tibell等2014、2019；Tuovila等2011、2013、2014；Vinuesa等2001；Vondrák等2022、2023；Vu等2019；Wang等2005；Wedin等2002、2005；Weerakoon等2012；Williams & Tibell 2008；Yahr 2015；Yang等2022。

系统发育分析：针对每个科与每个基因位点，本研究通过Mesquite软件（版本3.7，Maddison & Maddison 1997–2021）调用MAFFT软件（版本7.49，Katoh等2019、2002；Katoh & Standley 2013）的G-INS-i算法完成序列比对。研究人员通过目视检查比对结果，并使用Mesquite的‘高亮轻微错位区域’功能手动调整比对结果。本研究在DeCifr平台上使用ITSx 1.1软件（Bengtsson-Palme等2013）将序列拆分为ITS区、小亚基区与大亚基区文件，以辅助序列与比对结果的核查。使用GBLOCKS软件（Castresana 2000）在Mesquite中识别比对模糊的区域，采用较宽松的筛选标准（比对块最小长度设为2，允许所有字符存在 gaps，侧翼位置的最小非gap序列数设为该位置类群的51%），并将这些区域从后续分析中剔除。同时剔除序列末端与内含子区域。在Mesquite中，将未包含剔除区域的各基因位点序列进行拼接。原始比对文件（含剔除区域）与最终比对文件均存档于Dryad数据库，序列凭证数据详见补充表S2。