Black flies are important in public health and play a significant role in the epidemiology of parasitic diseases of humans, domesticated animals, and wildlife. Certain species of the genus Simulium are vectors of Onchocerca volvulus, the sole causative agent of human onchocerciasis. This disease infects more than 15 million people in Africa, Yemen, and Latin America, with approximately a million having lost sight (WHO 2022). In Southeast Asia, several pest species or species complexes have been reported, particularly in Thailand: Simulium asakoae, S. chamlongi, S. doipuiense complex, S. tenebrosum complex, S. umphangense, S. nigrogilvum, S. nodosum, S. khelangense, and S. chumpornense (Saeung et al 2020). Some of these black flies transmit filarial nematodes, such as Onchocerca spp., and blood protozoa, such as Leucocytozoon spp., among domestic animals and wildlife (Fukuda et al. 2003; Ishii et al. 2008; Jumpato et al. 2019; Pramual et al. 2020; Saeung et al. 2020; Takaoka et al. 2003; Thaijarern et al. 2019).

Baseline taxonomic information is available for many black flies in Southeast Asian countries (Takaoka et al. 2017a, 2017b, 2017c, 2018a, 2019), and their biodiversity has been documented in Indonesia, Malaysia, Thailand, and Vietnam (Adler et al. 2016; Takaoka and Davies 1995; Takaoka et al. 2015, 2018b; Hadi and Takaoka 2018). Currently, approximately 96 species of black flies have been recorded in Malaysia (Adler, 2022). However, the specific biting and vector species remain unknown. Additionally, there is insufficient molecular information available for many species, despite the availability of robust DNA barcode databases in neighboring countries such as Thailand, Vietnam, and Indonesia (database (Pramual et al. 2021; Hew et al. 2023; Putt et al. 2023). Ensuring accurate identification of Simuliidae is crucial for future public health research in this region, including Malaysia. By eliminating doubts of misidentification, researchers can confidently conduct studies and make informed decisions to address potential health concerns related to these insects. A reliable reference library for species ensures that each barcode can be used globally for identifications and comparisons (Ratnasingham and Hebert 2007; Sayers et al. 2020). In this case, type specimens or specimens collected from type localities (topotypes) are the most reliable for establishing the reference database (Kvist et al. 2010).

In the workflow application of DNA barcoding, organisms need to be identified and verified by taxonomists before DNA sequences can be deposited in the reference library (Hajibabaei et al. 2007; Janzen et al. 2005). This procedure ensures the reliability of the sequences for other researchers, especially non-taxonomists. With the combination of high-quality taxonomic descriptions and a strong reference library of specific taxa, DNA barcode applications can be swiftly performed with ease (Hajibabaei et al. 2006; Hebert et al. 2004; Janzen et al. 2005). Black flies are ideal candidates for DNA barcoding because they have been taxonomically studied throughout the world (Adler et al. 2004; Takaoka 2012, 2017; Hernández-Triana et al. 2015, 2017) and the taxonomy and distribution per country of each species have been summarized (Adler 2022). A molecular database for black flies has been developed in selected areas of the Nearctic Region (Rivera and Currie 2009), Mesoamerican countries (Hernández-Triana et al. 2015), Europe (Hernández-Triana et al. 2017; Ruiz-Arrondo et al. 2018), Thailand (Pramual et al. 2016, 2021; Pramual and Adler 2014), Indonesia (Hew et al. 2023) and Vietnam (Putt et al. 2023). To fill the knowledge gaps of species boundaries, the present study aims to establish a complete DNA barcode reference for Malaysian black flies, using topotypes and verified voucher specimens.

An increasing demand for accurate and timely species identification, and rapid advances in genetic methodology, have spurred progress in establishing molecular identification tools for black flies. By using short standardized DNA markers from mitochondrial COI, molecular species identification is becoming standard practice in many regions, world wide. The increasing importance of black flies in Malaysia makes the development of comprehensive and reliable species databases essential for accurate identification and comparison of regional faunas. A total of 52 species of black flies from Malaysia are represented in our study, accounting for more than 50% of the total species recorded in the country (Adler 2022) and including newly generated sequences for 14 species. The remaining species are unavailable, mostly owing to their rarity. Yet, their barcodes could be available in the future when more specimens are found or through non-destructive DNA extraction from museum specimens.

The distance-based approach successfully identified almost 90% of the species, in parallel with morphological identification, except for several species considered ambiguous or misidentified due to overlapping genetic distance with the nearest-neighbour species and high intraspecific distance. Despite the small number of samples per species, the genetic divergence overlap in this study (8.38%) is less than in the comprehensive study of black flies in Thailand (12.6%) (Pramual et al. 2021).

The barcode gap is the difference between intra- and interspecific genetic distances for each species (Robinson et al. 2009). A larger gap reflects greater accuracy of the DNA barcoding identification (Meyer and Paulay 2005). Absence of this gap leads to failed species separation due to overlap with the nearest-neighbour species. In certain cases, genetic overlap is due to the species having large genetic diversity (DeSalle et al. 2005) or closely related species with incomplete lineage sorting that may lead to non-monophyly (Funk and Omland 2003). In our study, almost 17% of the species lacked barcode gaps, and the sequences were identified as ambiguous after a failure to differentiate species. This result might be due to inadequate mitochondrial signals (Low et al. 2015) and incomplete lineage sorting (Low et al. 2016a). The nuclear 28S gene, however, can resolve some relationships, for example, between S. lurauense and S. sofiani but not of other members of the S. asakoae species group (Low et al. 2015), supporting the use of multi-locus genes as an alternative for distinguishing species rather than using a COI gene alone. Yet, the COI gene is still the gold standard for molecular identification of black flies. It has been used extensively in public databases for comparison.

A genetic distance for black fly species of more than 3% suggests the presence of cryptic species (Pramual and Adler 2014; Rivera and Currie 2009). By this metric, possible cryptic species are present in S. auratum and S. pegalanense, where the mean intraspecific differences are high (> 4%) and the species are placed in different lineages of the DNA barcode tree. Morphological re-examination is needed after a population study is done with more samples from various geographic distributions. Cryptic and pseudocryptic species are common in black flies, the former referring to two or more morphologically indistinguishable taxa within a nominal species (Bickford et al. 2006) and the latter referring to taxa that are distinguishable after thorough morphological examination coupled with other diagnostic techniques such as population genetics and cytogenetics (Low et al. 2014, 2016a, 2016b; Pramual and Kuvangkadilok 2012).

In the subgenus Nevermannia, S. aureohirtum in Malaysia shows high divergence (> 7%) from populations in Thailand. Delimitation of both species shows them as different entities. This generalist species has a wide geographic distribution, which could affect the species genetically. Chromosomal and molecular studies show differences among populations (Pramual et al. 2008; Thaijarern et al. 2014). Hence, morphological re-examination is suggested to evaluate the possibility of two separate entities. For the S. feuerborni species group, the low genetic difference between S. ledangense and S. pairoti is similar to the results of the study by Ya’cob et al. (2017a). However, distinct morphological characters are present between these two species and S. pairoti and S. ledangense are considered pseudocryptic species (Ya’cob et al. 2017a).

In the subgenus Simulium, S. kiuliense of the S. nobile species group is in the same lineage with species of the S. melanopus species group (S. mirum and S. murudense). Even species delineation indicates they are a single entity. Simulium kiuliense was revalidated after re-examination of S. nobile s. l., which revealed morphological differences and a large molecular distinction between populations in Java and mainland Asia along with a third species, S. vanluni (Low et al. 2016a; Ya’cob et al. 2017b). Simulium kiuliense is known only from the island of Borneo. The low interspecific divergence between these species might be related to an island effect, that is, low genetic diversity due to genetic drift and a possible bottleneck (Boessenkool et al. 2007; Eldridge et al. 2004). This possibility is in line with the genetic distances found in our study, whereby a large genetic distance (> 11%) exists within the S. nobile species group between S. kiuliense and S. vanluni but not (< 2%) between S. kiuliense and two species of the S. melanopus species group (S. mirum and S. murudense). Despite the high genetic similarity, morphological differences are present between the S. nobile and S. melanopus groups (Takaoka 2017).

Subgenus Gomphostilbia is the largest Simuliidae taxon in Malaysia, consisting of 50 species. In total, 58% of the Gomphostilbia species known from Malaysia were available for our study. In the S. batoense species group, S. parahiyangum and S. sazalyi are in the same lineage and are considered one entity due to incomplete lineage sorting (Sriphirom et al. 2014). This similarity is a consequence of one of four populations of S. sazalyi analysed from Peninsular Malaysia, which shares a COI sequence with S. parahiyangum from Sarawak (Takaoka et al. 2018c). Yet, these two nominal species can be separated morphologically in each life stage above the egg. On the other hand, sequences from seven of eight species in the S. asakoae species group are correctly identified in our study, with the exception of S. lurauense and S. sofiani, which are ambiguous and molecularly misidentified. All species in the S. epistum species group show a DNA barcode gap, except S. cheongi and S. whartoni, which are considered ambiguous due to high sequence similarity. Even the morphological characteristics of these species are similar (Takaoka and Davies 1995). Simulium whartoni is widely distributed from low to high elevations, but S. cheongi has been found only at low elevations (Takaoka and Davies 1995; Ya’cob et al. 2016). All insular species known only from East Malaysia in the S. epistum species group (S. auratum, S. barioense, S. kelabitense, and S. sarawakense) are correctly identified in our study and distinguishable from mainland species in the DNA barcode tree. This distinction is supported by the large genetic distance (> 13%) between mainland and insular species, suggesting that no genetic migration had occurred.

Overall, 22% of the species in our study are non-monophyletic and almost half of their sequences are ambiguous at the species level. This finding is consistent with that of Funk and Omland (2003), who reported nearly 26% of arthropods as non-monophyletic due to incomplete lineage sorting, lack of phylogenetic information, and imperfect taxonomy. These molecular issues are common in black fly studies (Low et al. 2016b; Sriphirom et al. 2014), although the species can often be differentiated morphologically. The remaining species in our study are correctly identified molecularly, suggesting that DNA barcoding of Malaysian black flies is acceptable with the support of species delimitation approaches. A good DNA barcode database can be established using specimens that have been identified morphologically before undergoing molecular analysis to increase accuracy for future referencing. A large sample size from various localities also will help increase the reliability of the species sequences (Meyer and Paulay 2005). The use of variable genetic markers, such as the fast-evolving nuclear big zinc finger (BZF) and elongation complex protein 1 (ECP1) genes, is also encouraged to improve identification efficiency, especially for problematic taxa (Aupalee et al. 2022; Low et al. 2016b; Pangjanda and Pramual 2017; Srisuka et al. 2022).

In summary, we report the DNA barcode for black flies in Malaysia, with high (90%) accuracy for species identification. An increase in the number of sequences per species deposited in DNA barcode databases, particularly when based on correct species identifications, will enhance the possible applications, such as monitoring vectors and other species of public health importance.

Noor Izwan-Anas, Van Lun Low and Zubaidah Ya’cob designed research and analyzed the data.

Zubaidah Ya’cob, Mohamad Rasul Abdullah Halim, Van Lun Low, Mohd Sofian-Azirun and Hiroyuki Takaoka collected the samples in the field.

Noor Izwan-Anas, Emmanuel Yogan Lourdes, Van Lun Low and Zubaidah Ya’cob performed the research.

Noor Izwan-Anas, Van Lun Low, Zubaidah Ya’cob, Hiroyuki Takaoka and Peter H. Adler wrote the paper.