Optimization of the real-time PCR platform method using high-resolution melting analysis and comparison with sequencing and phylogenetic analysis for developing optimal malaria diagnostic methods

Abstract

Malaria significantly impacts public health in developing countries. Accurate identification of Plasmodium species, which can cause chronic infections lasting months or years, is crucial for reducing infection rates. Despite the global availability of various diagnostic methods, challenges remain compared to the microscopic slide detection method. Molecular techniques, while more sensitive, are generally more expensive than microscopy but offer greater accuracy and automation potential in clinical settings. This study aims to optimize the real-time PCR platform method using high-resolution melting (HRM) analysis technology and compare the results with sequencing and phylogenetic analyses to develop optimal malaria diagnostic methods. A total of 300 samples were collected from individuals with suspected malaria symptoms in various regions of Sistan and Baluchistan province, located in southeastern Iran. Peripheral blood slides were examined microscopically and analyzed using PCR HRM with the Light Cycler 96 Instrument (Roche). Subsequently, phylogenetic analysis was conducted to further analyze the sequencing, with the desired sequences submitted to GenBank. Out of the 300 suspected samples, the PCR method identified 9 cases (3%) of Plasmodium falciparum and 20 cases (6.66%) of Plasmodium vivax. In comparison, the HRM method detected 15 cases (5%) of Plasmodium falciparum and 14 cases (4.66%) of Plasmodium vivax. Sequencing results revealed 13 cases (4.33%) of Plasmodium falciparum and 16 cases (5.33%) of Plasmodium vivax. The HRM method targeted the 18S SSU rRNA region, achieving a significant differentiation of 2.73 degrees to distinguish between the two species. The results confirm that, with proper primer design and precise species separation, the HRM technique is a reliable and optimal method for diagnosing malaria species. Also, The HRM method showed high sensitivity and specificity in identifying Plasmodium species, with complete agreement observed with sequencing in the tested samples.

Introduction

Malaria is a widespread infectious disease characterized by a wide spectrum of clinical manifestations, ranging from asymptomatic carriers to severe acute infections¹. According to the World Health Organization (WHO), approximately 229 million malaria cases are reported globally each year, with significant morbidity and mortality, particularly among children under the age of five and pregnant women². Over 120 species of malaria parasites are capable to infect a wide range of hosts, including mammals, reptiles, and birds. Among these, several subspecies of Plasmodium can infect humans, importantly Plasmodium falciparum, Plasmodium vivax, Plasmodium malariae, and Plasmodium ovale, which consists of two genetically distinct subspecies: P. ovale curtisi and P. ovale wallikeri^3,4. Infections caused by P. vivax and P. falciparum can persist for several months to years in human host⁵.

Accurate and timely diagnosis is essential for effective malaria management, as delays or mistakes in diagnosis can lead to disease progression, complications, and further transmission. Traditionally, microscopic examination of Giemsa-stained blood smears has been considered the gold standard for malaria diagnosis⁶. This method allows for species identification, evaluation of parasite developmental stages, morphological analysis, and quantification of parasitemia at a relatively low cost. Nevertheless, limited sensitivity, detecting only 10–50 parasites per microliter of blood, making it insufficient for identifying asymptomatic or low-level infections^7,8.

Various molecular methods such as nested PCR, quantitative real-time PCR (qPCR), and sequencing have been widely applied for the detection and species identification of Plasmodium parasites. Among these, high-resolution melting (HRM) analysis has emerged as a rapid, closed-tube, and cost-effective alternative, enabling species differentiation based on melting temperature (Tm) profiles of amplified DNA targets^9,10. PCR, in particular, is commonly used due to its ability to detect low parasite densities (0.02 parasites per microliter). Real-time PCR enhances sensitivity, reportedly detecting as low as 0.02 parasites/µL of peripheral human blood under controlled laboratory conditions¹¹. This detection limit may vary depending on the source of blood (e.g., peripheral vs. venous) and storage medium (e.g., dried blood spots). However, the widespread use of real-time PCR has been limited by high cost and the need for specialized equipment and technical expertise.

HRM analysis, a newer technique, offers a rapid screening opportunity and has emerged as a valuable tool for accurate species identification and detection of genetic polymorphisms in Plasmodium species^12,13. HRM offers several advantages, including its simplicity, speed, and cost-effectiveness. It has shown promise in the identification of single-nucleotide polymorphisms (SNPs) and differentiating species in various pathogens, including Plasmodium¹⁴. Previous studies have demonstrated the utility of HRM in malaria diagnosis. Rojas et al. reported 97.49% concordance between HRM and sequencing¹⁰. Chua et al. compared HRM with the commercial PlasmoNex™ system and highlighted its effectiveness in differentiating Plasmodium species based on 18S rRNA gene polymorphisms¹⁵. When combined with real-time PCR platform, HRM has the potential to enhance diagnostic accuracy without the need for sequencing¹⁶.

Despite its advantages, HRM remains underutilized in malaria research and diagnostics. This study employs HRM analysis for the detection and differentiation of Plasmodium species in blood samples collected from various regions of Sistan and Baluchistan province, Iran. Furthermore, the study aims to optimize real-time PCR platform protocols using HRM technology, and compare the results with conventional DNA sequencing and phylogenetic analysis.

Materials and methods

Clinical sample collection

Samples were collected in 2019 from febrile individuals suspected of malaria who presented to hospitals and primary health care centers in Sistan and Baluchestan Province, southeastern Iran. The sample size (n = 300) was determined based on previous studies in the region reporting a malaria prevalence of approximately 3% and an estimated diagnostic sensitivity of 58%. This number was considered sufficient to detect asymptomatic infections with acceptable statistical power and logistical feasibility. After conducting Rapid Diagnostic Test (RDT), blood samples were collected from the patients. Thick and thin blood smears were prepared, and DNA extraction was performed in the Department of Parasitology at Mashhad University of Medical Sciences.

Ethical approval

The protocol approval of this study was obtained from the Ethics Committee of Mashhad University of Medical Sciences on January 30, 2019 (IR.MUMS.MEDICAL.REC.1397.601). All the protocols of our study were performed following the ethical standards of the institutional research committee and the Declaration of Helsinki. Before sample collection, all participants were comprehensively informed about the study, and signed the written informed consent. In the children between 12 and 18 years of age, assent was also obtained from their parent/legal guardian. Patients’ information was kept confidential. All the patients voluntarily entered in our study and their participation did not affect the provision of treatment at the respective health facilities.

Geographical context

Sistan and Baluchistan Province, located at 27.5300° N and 60.5821° E, covers a vast area of nearly 180,726 square kilometers (Fig. 1), comparable to Syria. This region features a desert climate and is still underdeveloped, with a low socio-economic condition. Additionally, it is well-known as an endemic region for malaria.

Fig. 1.

Sistan Baluchistan province (southeast) and its studied areas.

Conventional microscopic examination and determination of parasitemia

Thin and thick blood smears stained with Giemsa were prepared, and experienced specialists examined them at 1000X magnification. Parasites were identified based on morphology and developmental stages (early trophozoite, late trophozoite, schizonts, and gametocytes). Parasitemia was quantified using a grading system: + (1–10 parasites/100 fields or 4–40 parasites/µL), + + (11–100 parasites/100 fields or 4–400 parasites/µL), + + + (1–100 parasites/field or 401–4000 parasites/µL), and ++++ (≥ 10 parasites/field or > 4000 parasites/µL). Negative samples after examining 200 fields were recorded as negative¹⁷.

Genomic DNA extraction and PCR assay

The first round of PCR was necessary to increase template availability for nested amplification, enhancing overall sensitivity and specificity, especially in samples with low parasitemia. Genomic DNA from all samples used in HRM, nested PCR, and sequencing was extracted using the Qiagen DNA Mini Kit (QIAGEN, Germany or Gent Bio PrimePrepTM), following the manufacturer’s protocol and stored at − 20 °C¹⁸. DNA concentration was determined using a NanoDrop spectrophotometer (Thermo OneC Microvolume UV–Vis, USA).

PCR amplification

PCR amplification targeting the conserved region of the 18S SSU rRNA gene was performed using primers MEH (Forward) and UNR (Reverse). PCR amplifications were carried out in a 20 μl reaction volume containing 1 × buffer, 2.5 mM MgCl₂, 200 μM dNTPs, 200 nM primers, and 1U Taq DNA-polymerase with approximately 10 ng of DNA template. It was performed using a BIO-RAD T100™ Thermal Cycler. Primers MEH (Forward)—5′-GAACGGCTCATTAAAAACAGT-3′ and (UNR)—5′-GACGGTATCTGATCGTCTTC-3′ were used to amplify Plasmodium DNA. The PCR conditions included an initial denaturation at 95 °C for 5 min, followed by 40 cycles of denaturation at 94 °C for 45 s, annealing at 60 °C for 45 s, extension at 72 °C for 70 s, and final elongation at 72 °C for 10 min. The sequences and product sizes of all primers used in PCR and nested PCR are listed in Table 1.

Table 1.

Primer sequences, specificities, and product sizes for PCR and nested-PCR amplification of Plasmodium species.

Primer	Sequence (5′–3′)	Specificity	Product size (bp)†
First reaction (PCR)
Forward primer
All	GAACGGCTCATTAAAACAGT	All species	923††
Reverse primer
UNR	GACGGTATCTGATCGTCTTC	Universal	*
Second reaction (nested-PCR)
Forward primer
All	GAACGGCTCATTAAAACAGT	All species
Reverse primer
OVR	GCATAAGGAATGCAAAGAACAG	P. ovale	436*
FAR	AGTTCCCCTAGAATAGTTACA	P. falciparum	395*
MAR	GCCCTCCAATTGCCTTCTG	P. malariae	269*
VIR	AGGACTTCCAAGCCGAAGC	P. vivax	499*

^†Bp (Base pairs).

^††Size depending upon species: P. malariae (821 bp), P. falciparum (787 bp), P. vivax (783 bp), and P. ovale (794 bp).

*(103).

Nested PCR

To increase the sensitivity of detection and allow species-level identification, a nested PCR approach was employed. A 1:1000 dilution of the first-round PCR product was prepared, and 3 µL of this diluted product was used as the template for the second-round reaction. The 20 µL nested PCR mixture contained 1 × buffer, 2.5 mM MgCl₂, 200 µM dNTPs, 200 nM of each species-specific primer (FAR, VIR, MAR, OVR), and 1 U of Taq DNA polymerase. These primers target P. falciparum, P. vivax, P. malariae, and P. ovale, respectively, as described by Rubio et al.¹⁹. The second-round PCR was conducted with an initial denaturation at 95 °C for 4 min, followed by 35 cycles of 94 °C for 20 s, 60 °C for 20 s, and 72 °C for 45 s, ending with a final extension at 72 °C for 10 min. PCR products were analyzed by 1% agarose gel electrophoresis run along with a 100 bp DNA ladder. Due to the small difference in band size between P. falciparum and P. vivax, identification was primarily confirmed by species-specific primers and melting profile, as a 1% agarose gel alone may lack the resolution to distinguish 4-bp differences.

Quantitative HRM assay development and primer design

The primers used in the HRM-real-time PCR assay (MEH and Omid) were designed in-house using Primer3 software, based on the alignment of conserved regions of the 18S SSU rRNA gene from Plasmodium species available in GenBank (accession numbers: L48987, M54897, X13926, M19172) (Fig. 2). These were validated through in silico BLAST analysis to ensure specificity²⁰. The primers were selected to amplify a fragment suitable for HRM analysis, providing specificity and compatibility with the EvaGreen dye-based detection system. Real-time PCR coupled with HRM was performed on a LightCycler® 96 instrument (Roche Diagnostics GmbH, Mannheim, Germany). Each 20 µL reaction contained 10 µL of 2X Hot FirePol EvaGreen HRM Mix (Solis BioDyne, Tartu, Estonia), 0.5 µM of each primer (MEH Forward: 5′-GAACGGCTCATTAAAAACAGT-3′ and Omid Reverse: 5′-GGAATTATAACAAAGAAGTAACAC-3′), 4 µL of template DNA (5–20 ng/µL), and nuclease-free water to adjust the volume. The thermal cycling method consisted of primary pre-incubation at 95 °C for 700 s, followed by 45 cycles of amplification (denaturation at 95 °C for 15 s, annealing at 52 °C for 20 s, and elongation at 72 °C for 20 s), and a last extension at 72 °C for 300 s. HRM analysis was conducted immediately post-amplification, with a melting curve generated from 60 °C to 97 °C at a ramp rate of 0.07 °C/s and continuous fluorescence acquisition (10 readings/°C). Melting profiles were analyzed using LightCycler® 96 software (version 1.1, Roche) to determine species-specific Tm based on derivative curves, aligned curves, and difference plots. An annealing temperature of 52 °C was selected based on gradient PCR optimization to ensure specific amplification across Plasmodium species while minimizing non-specific binding. EvaGreen was chosen due to its high sensitivity and compatibility with HRM analysis, offering low background fluorescence and high resolution in melting profile differentiation. Each HRM-PCR run included positive controls (sequencing-confirmed P. falciparum and P. vivax samples) and a no-template negative control. All reactions were performed in three technical replicates to ensure reproducibility and accuracy. In addition, all HRM optimization and validation experiments were conducted using the same set of clinical samples to ensure consistency in evaluation.

Fig. 2.

Schematic map of four different species of Plasmodium; The position of primers PCR and Nested-PCR marked with different colors: primers ((MEH and UNR PCR) red arrow), Nested-PCR primers (FAR, MAR, VIR, and OVR) with the green arrow, HRM primer (blue arrow), (Chun article primer in yellow color) and rRNA (black and brown arrow) have been identified using CLC bioinformatics software.

Phylogenetic analysis

PCR products were purified using the QIAquick PCR Purification Kit (Qiagen, Hilden, Germany) and sequenced bidirectionally with the same primers (MEH and Omid) on an ABI 3500 Genetic Analyzer (Applied Biosystems, Foster City, CA, USA). Sequence chromatograms were inspected, trimmed, and aligned using Sequencher v4.4 (Gene Codes Corporation, Ann Arbor, MI, USA) and CLC Main Workbench v12 (QIAGEN, Aarhus, Denmark). Consensus sequences were compared against the NCBI database using BLASTn to confirm species identity. Although the 18S SSU rRNA gene is conserved, it contains sufficient interspecies variability for phylogenetic clustering and has been widely used in Plasmodium genotyping studies^21,22. The evolutionary history was inferred by using the Maximum Likelihood method based on the Tamura-Nei model²³. The tree with the highest log likelihood (-596.33) is shown. The percentage of trees in which the associated taxa clustered together is shown next to the branches. Initial tree(s) for the heuristic search were obtained automatically by applying Neighbor-Join and BioNJ algorithms to a matrix of pairwise distances estimated using the Maximum Composite Likelihood (MCL) approach, and then selecting the topology with superior log likelihood value. The analysis involved 27 nucleotide sequences. All positions containing gaps and missing data were eliminated. There was a total of 361 positions in the final dataset. Evolutionary analyses were conducted in MEGA7²⁴. Toxoplasma gondii (GenBank Accession: X75429) was included as an outgroup to root the tree. Sequences generated in this study were deposited in GenBank under accession numbers MT705725—MT710335 and MT710323—MT710339 for Plasmodium falciparum and Plasmodium vivax, respectively (Plasmodium vivax) (https://www.ncbi.nlm.nih.gov/nuccore/?term=(zarean)+AND+%22Plasmodium+vivax%22%5Bporgn%3A__txid5855%5D) and (https://www.ncbi.nlm.nih.gov/nuccore/?term=(zarean)+AND+%22Plasmodium+falciparum%22%5Bporgn%3A__txid5833%5D).

Statistical analysis

Melting temperatures (Tm) were determined from three technical replicates per sample, and means ± standard deviations (SD) were determined using LightCycler® 96 software. The coefficient of variability (CV) was computed as CV = (SD/mean Tm) × 100 to evaluate assay reproducibility. Differences in Tm between Plasmodium species were assessed for statistical significance using a one-way assessment of variance (ANOVA) followed by Tukey’s post-hoc test in SPSS v22 (IBM Corp., Armonk, NY, USA). The sensitivity and specificity of the HRM assay were compared against nested PCR and sequencing results, with microscopy serving as the reference standard.

Results

Targeted region and primer design

The conserved region of 18S SSU rRNA was targeted using the forward primer MEH and a newly designed reverse primer (Omid) (Table 2, Fig. 3). Blast analysis indicated that the amplicon covered regions less than 200 base pairs, facilitating differentiation among Plasmodium species in melting curve analysis. Specificity was confirmed by absence of non-specific products during PCR, with an optimal annealing temperature of 60 °C selected for HRM assays. To support HRM primer design, we performed in silico analysis of conserved 18S SSU rRNA gene sequences retrieved from GenBank. Details of the sequence information used for HRM primer design are presented in Table 3. Table 4 summarizes key features of these sequences, including their predicted melting temperatures under various salt concentrations, which guided the selection of appropriate primer targets and expected HRM profiles.

Table 2.

List of HRM primers.

Primer	Sequence (5′-3′)	Specificity
Reaction (HRM)
Forward primer
All (MEH)	GAACGGCTCATTAAAACAGT	All species
Reverse primer
Omid	GGAATTATAACAAAGAAGTAACAC

Fig. 3.

Schematic map of the location of the newly designed Omid primer for targeting the 18S SSU rRNA region in HRM in four different species of Plasmodium: (MEH and Omid (red and blue arrows), primers from Chua et al.’s article (yellow arrow) and rRNA (black and brown arrow) using bioinformatics software CLC Genomic Workbench12.

Table 3.

Sequence information.

Information	PFARGEA	PFARGBAB	PFASSURA	X13926
Sequence	DNA	DNA	DNA	DNA
Length	153bp	157bp	159bp	152bp
Organism	Plasmodium falciparum	Plasmodium malariae	Plasmodium ovale	Plasmodium vivax
Name	PFARGEA	PFARGBAB	PFASSURA	X13926
Description	P. falciparum 18S ribosomal RNA in asexual parasites	P. malariae small subunit ribosomal RNA gene	P. ovale small subunit ribosomal RNA gene complete sequence	P. vivax small subunit (SSU) rRNA gene
Modification date	26-APR-1993	26-APR-1993	22-OCT-1997	13-JUL-1998
Weight (single-stranded)	47.318 kDa	48.535 kDa	49.113 kDa	46.896 kDa
Weight (double-stranded)	94.54 kDa	97.009 kDa	98.254 kDa	93.926 kDa

Table 4.

Melting temperatures—degrees celsius.

(Salt)	PFARGEA	PFARGBAB	PFASSURA	X13926
0.1 M	72.89	72.68	75.16	74.02
0.2 M	77.88	77.68	80.16	79.02
0.3 M	80.81	80.60	83.08	81.94
0.4 M	82.88	82.68	85.16	84.01
0.5 M	84.49	84.29	86.77	85.62

Melting temperature (Tm)

HRM analysis successfully identified Plasmodium species based on distinct melting curves (derivative curve, aligned curve, and difference plot). The melting temperatures (Tm) were determined as follows: P. vivax (79.02 °C), P. falciparum (77.88 °C), P. malariae (77.68 °C), and P. ovale (80.16 °C) (Fig. 4). These findings were in agreement with the in-silico predictions shown in Table 4, which support the validity of targeting this region for species differentiation (Fig. 5).

Fig. 4.

In-silico high resolution melting (HRM) analysis prediction by CLC DNA workbench software to determine the melting point of each reference strain. 1.1 The detail of Malaria species based on Gen bank contents. 1.2: T_m predictions of HRM analysis in malaria species.

Fig. 5.

Melting curve variance of asymptomatic malaria in (A) normalized Melting curve (B) derivative and (C) difference plot analyses. The melting curves and Tm for each species can be very well understood. Red = P. falciparum, Blue = P. vivax, Black = control (no template).

Assay specificity and sensitivity of HRM

HRM using MEH-F and Omid-R primers demonstrated high specificity, correlating well with species-specific control plasmids. Changes in cycle threshold (CT) indicated successful PCR amplification. The method effectively distinguished between P. vivax and P. falciparum, showing slight similarity with Toxoplasma gondii species. Minor similarity observed in silico between Plasmodium primers and Toxoplasma gondii sequences was negligible, with no cross-reactivity observed in experimental runs, confirming high assay specificity.

Estimation of melting temperatures, standard deviation, and coefficient of variation

Analysis of Tm values showed a 2.73 °C difference between P. falciparum and P. vivax (Table 5), highlighting the discriminatory power of HRM in species identification. Species discrimination was based on characteristic Tm, with interspecies differences of ≥ 1.5 °C considered sufficient for clear differentiation. These thresholds were determined through analysis of reference controls and sequencing-confirmed samples.

Table 5.

Mean Tm, SD, and CV calculated based on of 18ssu rRNA gene sequence of P. falciparum and P. vivax.

Gene	Mean TM (°C)	SD	CV
P. falciparum	77.08 °C	0.11	0.15
P. vivax	79.81 °C	0.17	0.21

SD, standard deviation; CV, coefficient of variation.

Comparison with sequencing and nested-PCR

Prior to molecular analysis, all 300 peripheral blood samples were initially screened microscopically as the gold standard. However, due to the known limited sensitivity of microscopy for low-parasitemia infections, PCR-based methods were employed for confirmation and species identification. Of the 300 samples, only 29 were positive by at least one molecular method (Table 6). Specifically, Nested-PCR detected 20 cases (6.6%) of P. vivax and 9 cases (3%) of P. falciparum, whereas HRM identified 14 cases (4.6%) of P. vivax and 15 cases (5%) of P. falciparum. No infections from P. malariae, P. ovale, or mixed infections were observed in either method. Despite the high performance of HRM for P. falciparum, its sensitivity for detecting P. vivax was slightly lower, as six P. vivax-positive samples identified by nested PCR were not detected by HRM. This discrepancy may stem from low parasite density in asymptomatic individuals, reduced amplification efficiency for P. vivax, or sequence variability within the target region¹³. These factors may affect melting curve clarity or Tm detection, especially in borderline cases. In such instances, complementary confirmation by sequencing is recommended. Compared to sequencing as the reference standard, HRM shows a sensitivity of 100% and specificity of 99.31%, surpassing Nested-PCR (sensitivity: 100%, specificity: 98.55%), particularly for rapid detection of acute malaria cases (Table 6(. Discrepancies between methods may be attributed to differences in sensitivity thresholds and template availability. Sequencing was used as the gold standard for confirmation, and primers MEH and Omid were employed for amplification before sequencing.

Table 6.

The comparison of nested PCR and HRAM methods for the accurate diagnosis of malaria species in suspected patients.

Method	P. falciparum	P. vivax	Mixed	Negative
HRM	15	14	0	271
Nested PCR	9	20	0	271
Sequencing	13	16	0	271

Phylogenetic analysis

Sequencing confirmed P. vivax (16 cases, 5.3%) and P. falciparum (13 cases, 4.3%) identities, with amplicon size verification on gel electrophoresis (Fig. 6). Alignments revealed insertion mutations and single nucleotide changes in both species. To evaluate the analytical sensitivity and reproducibility of the HRM assay, serial dilutions of DNA from confirmed Plasmodium samples were subjected to PCR-HRM. As shown in Fig. 7, clear amplification and species-specific melting curves were obtained across the dilution series, indicating that the method can reliably detect low DNA concentrations without loss of species discrimination.

Fig. 6.

Nested-PCR analysis of 18SSU rRNA gene on gel electrophoresis. Line 1,3,4,5: P. falciparum, line 2: Plasmodium vivax, line 6: positive control and line 7: negative control. Ladder: 100 bp Marker.

Fig. 7.

Representative amplification plot (A) and derivative melting curves (B).

Comparison with NCBI database

Sequenced samples exhibited high similarity with NCBI database entries, confirming species identification. Phylogenetic analysis using the Maximum Likelihood method aligned isolates from Sistan and Baluchistan province with global Plasmodium species. The phylogenetic tree showed that the studied isolates were most closely related to regional isolates of P. vivax and P. falciparum. The cluster structure of the tree also confirmed the accuracy of the species determination. The phylogenetic relationships between the isolates and reference sequences are shown in Fig. 8.

Fig. 8.

Phylogenetic tree constructed based on partial sequences of the 18S SSU rRNA gene. The tree includes Plasmodium falciparum and Plasmodium vivax isolates from this study and reference sequences retrieved from GenBank. Bootstrap values above 1% are shown at the nodes.

Discussion

Malaria is a global health concern, particularly in resource-limited areas where effective diagnosis plays an important role in disease management and control efforts. Our study aimed to optimize the real-time PCR method using HRM analysis and compare it with traditional methods for malaria diagnosis. The results show that HRM technology offers significant advantages in terms of accuracy, speed, and cost-effectiveness compared to conventional techniques.

Significance of the optimized HRM method

There is a 2.73 °C temperature difference in melting points (Tm) between P. falciparum (77.08 °C) and P. vivax (79.81 °C) seen in our study, a key finding that shows the power of our optimized HRM technique. Chua et al. revealed a 1.5 °C temperature difference between two species in their study¹⁵, revealing improved resolution in species differentiation. The alignment of experimental melting temperatures with in silico predictions (Table 4) underscores the robustness of our HRM primer design. The enhanced accuracy can be attributed to our careful primer design targeting the variable regions of the 18S SSU rRNA gene, which includes acceptable sequence variations to distinguish between Plasmodium species while maintaining appropriate amplicon length for optimal HRM performance. HRM provided species-specific Tm values with high reproducibility, as reflected by the low intra-species variation observed in Table 5. This reproducibility reinforces the method’s reliability in distinguishing Plasmodium species. Minor variations in Tm values were observed among some samples of the same Plasmodium species, as shown in Fig. 5. These differences may reflect natural sequence polymorphisms within the target 18S rRNA region or slight technical variations such as DNA concentration or PCR conditions. However, the intra-species Tm variation remained limited (as shown in Table 5), and species-level discrimination was unaffected.

Our HRM method’s sensitivity and specificity values (100% and 99.31%, respectively) are slightly improved than those of similar molecular techniques. For instance, Kipanga et al. reported sensitivity and specificity values of 97% and 96% for their nested PCR-HRM assay²⁵. The high accuracy of our method resulted from optimized reaction conditions and the targeting of species-specific regions within the 18S SSU rRNA gene, enabling precise discrimination between Plasmodium species. While our HRM method demonstrated high resolution, we acknowledge that the Tm difference between P. falciparum (77.88 °C) and P. malariae (77.68 °C) was very narrow (~ 0.2 °C), which could pose challenges under field conditions with limited instrument precision. However, under our controlled laboratory conditions, the HRM system consistently distinguished these species using both melting profiles and difference plots. In borderline cases, we recommend confirming results with sequencing to ensure diagnostic accuracy.

Several previous studies have applied HRM for Plasmodium species identification with varying protocols. Chua et al. used HRM targeting the 18S rRNA gene and reported 100% concordance with the PlasmoNex™ commercial kit, while our study showed comparable diagnostic accuracy with sequencing¹⁵. Kipanga et al. reported 97% sensitivity using nested PCR-HRM, slightly lower than our observed sensitivity of 100%, likely due to differences in primer design and amplicon length²⁵. Unlike these studies, we employed newly designed primers (MEH and Omid) with optimized annealing temperatures and minimal cross-reactivity, targeting conserved yet variable regions of the 18S SSU rRNA gene. Moreover, while it was validated HRM on cultured isolates or reference strains, our study applied the method to clinical samples, which may better reflect real-world diagnostic challenges¹³. These methodological distinctions may explain the enhanced Tm separation and overall performance of our protocol.

Clinical implications

In Sistan and Baluchistan provinces and other regions where both P. falciparum and P. vivax are prevalent, rapid and accurate identification of P. falciparum is critical due to its potential to cause severe and fatal malaria. Our HRM method provides results in approximately 2–3 h, which is faster than nested PCR (6–8 h) and comparable to real-time PCR (~ 2 h), with enhanced accuracy. Still, it’s not as accurate as HRM PCR (75% specificity to 99.31%), which is important for clinical decision-making and treatment initiation²⁶. This time advantage becomes vital in severe malaria cases where quick treatment reduces mortality rates.

Furthermore, the ability of HRM to detect six additional P. falciparum cases that were negative by nested PCR highlights its outstanding sensitivity in detecting low parasitemia infections. Murillo and Co-workers also analyze HRM as a new method for detecting low parasitemia infections in plasmodium types, and they stated that it could be accurate in detecting low parasitemia infections like nested PCR methods of Snounou et al., although our study delivers a better performance than nested PCR in detecting low parasitemia infections^13,26,27. As we know, missed diagnoses of low-density infections can contribute to persistent transmission cycles and undermine malaria elimination efforts²⁸; therefore, detection of low parasitemia infections has important implications for both individual patient management and broader public health initiatives.

Comparison with sequencing-based methods

As we know, sequencing-based approaches are highly accurate and remain the gold standard for detecting some infections like Plasmodium species²⁹. Nevertheless, they are often impractical for routine diagnostic use in endemic settings due to their high cost and technical complexity³⁰. In our study, HRM had excellent concordance in accuracy and species identification with sequencing-based methods. Furthermore, it does not have sequencing problems like expressivity and complexity. Rojas et al. demonstrated that HRM had 97.49% concordance with sequencing¹⁰, suggesting that our optimized HRM method could be a suitable replacement, particularly in resource-limited areas. Besides, our data indicated that Cohen’s kappa between HRM and sequencing was 0.999, reflecting almost perfect diagnostic agreement.”

Advantages over nested PCR

Nested PCR has been widely used for molecular diagnosis of malaria but has several limitations like risk of Contamination, consumption, cost, and equipment requirements; unlike nested PCR, HRM minimizes the risk of contamination due to its closed-tube system and may be faster and more cost-efficient than nested PCR, although we did not conduct a direct cost/time comparison^15,31. Furthermore, we explain that the HRM had better sensitivity and specificity to nested PCR. Anthony et al. comprise two nested PCR methods for the detection of human malaria, and they stated that 18S SSU rRNA primers and dhfr-ts primers had 91.9% and 51.4% sensitivity in order and 100% specificity; we can see our HRM methods with 100% sensitivity had better performance to both of nested PCR methods but its specificity (99.31%) is a little lower than Anthony study³²; it’s justified by seeing the gold standards for its research, microscopy detection of Plasmodium, which in our study was sequencing and is known to be more accurate, and the microscopic method may not have identified some samples³³.

Our findings revealed that the HRM method detected 15 cases of P. falciparum compared to 9 cases by nested PCR, show that HRM may find parasites in lower concentration. This aligns with research by Murillo that after comparison of HRM with other methods stated, HRM can detecting down to 1 parasite or 6 copies/μL but it is down to 10 parasites/µL for nested PCR method^13,34.

Comparison with commercial kits

PlasmoNex™ is a commercial kit for detecting Plasmodium species, with high reported accuracy and 100% sensitivity and specificity based on sequencing results³⁵. Chua et al. developed an HRM-based method similar to ours and reported complete concordance between HRM and PlasmoNex™¹⁵. Although we did not perform a direct comparison with commercial kits in our study, HRM protocols such as ours involve fewer post-PCR steps and closed-tube detection, which may offer practical advantages such as reduced turnaround time in some settings. These potential benefits, however, should be interpreted cautiously in the absence of formal cost or time comparisons^15,35,36.

Study limitations

Despite the promising results, our study has several limitations that should be acknowledged.

First, this study was limited by the low number of positive samples (29 out of 300), particularly for P. falciparum, which constituted only 3–5% of cases. This limited representation may affect the generalizability of our findings regarding HRM performance for this species. Therefore, caution is warranted in extrapolating the results to broader populations, and larger-scale studies are recommended.

Second, our sample is limited to a specific geographic region (Sistan and Baluchistan province), which may affect the generalizability of our results. We know parasite variation could influence the functionality of molecular diagnostic methods³⁷.

Third, while our study focused on the two most common species (P. falciparum and P. vivax), we didn’t consider less common species like P. malariae or P. ovale, so the reliability of HRM may need proof by additional samples of other species.

Finally, although no mixed infections were identified among the samples, we did not include spiked or simulated co-infection controls to assess the HRM method’s capacity to detect mixed species. Future validation with artificial mixtures is recommended. While we explained the analytical performance of our HRM method, it needs more assessment and field evaluation under real-world conditions to be introduced as a reliable method for malaria control programs worldwide.

Conclusion

In conclusion, our optimized HRM method is reliable and potentially more rapid and cost-effective than nested PCR due to its closed-tube format and simplified workflow. Compared to sequencing, HRM’s reliable performance validates its accuracy, and its operational advantages over nested PCR make it more suitable for routine use.

Because accuracy, turnaround time, and accessibility are critical in malaria management, HRM may offer a valuable complementary tool for improving diagnostics, especially in settings with limited resources. Further investigation and evaluation of this method could help control and eradicate malaria.

Author contributions

Omid Ahmadi contributed to study design, data collection, analysis, and manuscript preparation. Yousef Sharifi participated in data collection and interpretation. Michael Saeed assisted with statistical analysis and manuscript drafting. Amirali Reihani contributed to experimental procedures, manuscript drafting, and data validation. Seyed Aliakbar Shamsian supervised the study design and methodology. Elham Moghaddas assisted with laboratory procedures and data acquisition. Hadi Mirahmadi contributed to literature review and manuscript editing. Soudabeh Etemadi participated in PCR optimization and sequencing procedures. Reza Fotouhi-Ardakani provided project supervision and funding acquisition. Mehdi Zarean supervised all aspects of the study, including methodology, data analysis, and manuscript revision. All authors reviewed and approved the final manuscript.

Funding

This research was supported by Mashhad University of Medical Sciences, Grant No. 970708. M.Z had a grant from Mashhad University of Medical Sciences.

Data availability

The datasets generated during the current study are available in the Genbank repository under following links (https://www.ncbi.nlm.nih.gov/nuccore/?term=(zarean)+AND+%22Plasmodium+vivax%22%5Bporgn%3A__txid5855%5D) and (https://www.ncbi.nlm.nih.gov/nuccore/?term=(zarean)+AND+%22Plasmodium+falciparum%22%5Bporgn%3A__txid5833%5D).

Declarations

Competing interests

The authors declare no competing interests.