key: cord-1021964-msdh5ep6
authors: Mohanty, Abhijeet; Fatrekar, Adarsh P.; Krishnan, Saravanan; Vernekar, Amit A.
title: A Concise Discussion on the Potential Spectral Tools for the Rapid COVID-19 Diagnosis
date: 2021-05-06
journal: Results Chem
DOI: 10.1016/j.rechem.2021.100138
sha: bad721bd5425837dac395a2ba1af0570dce83f82
doc_id: 1021964
cord_uid: msdh5ep6

Developing robust methods to detect the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), a causative agent for the current global health pandemic, is an exciting area of research. Nevertheless, the currently used conventional reverse transcription-polymerase chain reaction (RT-PCR) technique in COVID-19 diagnosis endures with some inevitable limitations. Consequently, the establishment of rapid diagnostic tools and quick isolation of infected patients is highly essential. Furthermore, the requirement of point-of-care testing is the need of the hour. Considering this, we have provided a brief review of the use of very recently reported robust spectral tools for rapid COVID-19 diagnosis. The spectral tools include, colorimetric reverse transcription loop-mediated isothermal amplification (RT-LAMP) and matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS), with the admittance of principal component analysis (PCA) and machine learning (ML) for meeting the high-throughput and fool-proof diagnostic platforms for the detection of SARS-CoV-2, are reviewed. Recently, these techniques have been readily applied to screen a large number of suspected patients within a short period and they demonstrated higher sensitivity for the detection of COVID-19 patients from unaffected human subjects.

7744 copies/mL and 968 copies/mL, respectively [6] . Poor LOD and low sensitivity of these RT-PCR kits led to improper diagnosis of viral infections in the pandemic situation, which eventually result in false-negative results. Given this milieu, new screening techniques are required to correctly identify infected patients from probable cases and control the spread of these viral infections.

In this review, we aim at introducing matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) based and reverse transcription loop-mediated isothermal amplification (RT-LAMP)-based diagnostic tools with the intervention of artificial intelligence (AI) and machine learning (ML) techniques for the rapid and accurate detection of SARS-COV-2 virus. The application of AI and ML automation in health maintenance starts with the development of the first ML system called MYCIN [7] , which was programmed to suggest antibiotics for the treatment of bacterial infected patients by using a data of 450 rules collected from a team of medical experts. Subsequently, several ML techniques have been employed as promising technology against various contagious (SARS [8] [9] [10] , EBOLA [11] , HIV [12, 13] ) and noncontagious (Cancer [14] , Diabetic [15] , Heart disease [16] , and Stroke [17] ) epidemic outbreaks.

These aforementioned pieces of shreds of evidence encourage researchers to face down the current epidemic using effective approaches of ML and AI technologies.

We believe that this review could provide essential information to researchers who have interests in developing MALDI-MS and RT-LAMP-based modern rapid tools for detection of the COVID-19, as well as pose a potential picture on the extensive use of these useful diagnostic methods for battling the current pandemic.

The research group from the University of Talca and Autonomous University of Chile reported a holistic approach for the detection of SARS-CoV-2 in nasal swabs using MALDI-MS and ML ( Figure 1 ) [18] . In this study, a total of 362 nasal mucous secretion swab samples collected from three different laboratories located in different countries were tested by the standard RT-PCR technique. Results showed that 211 nasal swab samples were COVID-19 positive and the remaining were negative. The same sets of samples were subjected to MALDI-MS and spectra generated were preprocessed to obtain an intensity matrix with identified peak. The most important aspect of this study was the identification of the correct peak which delineates the information whether samples being handled were COVID-19 positive or negative. Further, feature selection (FS)-information gain (Ig)-based and correlation-based (Cfs) tools were applied to obtained intensity matrices to establish the signature peaks with significant differences between COVID-19 positive and negative mean spectra. All the obtained data were individually processed for principal component analysis In a separate study, a mass spectrometry (MS)-based approach was employed to detect the presence of viral proteins in a gargle solution of COVID-19 patients [19] . The key constituents present in the gargle solution are acetone and trypsin which were used for precipitation and digestion of proteins, respectively. MS analysis identified the presence of unique peptide fragments originated from SARS-CoV-2 nucleoprotein based on their mass/charge (m/z) value, which forms the assay principle for the clinical diagnosis of COVID-19.

Another study has been reported by Rocca et al. wherein they have demonstrated a potential tool for the diagnosis of COVID 19 by combining, matrix-assisted laser desorption ionization timeof-flight mass spectrometry (MALDI-TOF MS) with multivariate ML algorithms [20] . The group has collected nasopharyngeal swab samples in 2 ml of saline solution and stored them at -20 o C.

They have recorded the MALDI spectra within a range of 2 kDa-20 kDa, and thereafter differentiated the SARS-COV-2 positive and negative sample peaks by using two software such as Flex analysis v3.4 and ClinPro Tools software v3.0. The performance of the combined tool has been estimated by evaluating, accuracy (67.6 %), sensitivity (71.7 %), specificity (61.7 %), positive prediction (60 %), and negative prediction (73 %) by using the ClinPro Tools.

Very recently, Yan et al. [21] have reported a serum peptidome profiling method based on MALDI-TOF MS for efficient and rapid detection of COVID-19. After processing the MS data, they have implemented 8 ML methods to construct the classification models based on which a very high accuracy rate of 99 %, with a specificity of 100 % and sensitivity of 98% have been achieved. They focused on the analysis within a mass range of 5000 to 30,000 m/z, which corresponds to the serum peptidome and small proteins. A very minute volume of sample (5 μL) has been analyzed within 1 min with a cost of < 1 USD per sample.

Loop-mediated isothermal amplification (LAMP) is a technology for nucleic acid amplification in 30 min, at one constant temperature (~ 65 0 C). This method has been employed for the last few years to detect viral and bacterial pathogen [22] [23] [24] . Usually, the LAMP method is used for DNA amplification (ref) in presence of 4 to 6 specially designed primers to bind corresponding regions of the target DNA with very high specificity [25, 26] . However, to detect RNA sequence, the LAMP protocol has been incorporated with the reverse transcription method and collectively called RT-LAMP technology. In the RT-LAMP method, the viral RNA has been converted to complementary DNA that undergoes amplification at isothermal conditions [27] . The assay involved in the RT-LAMP method is colorimetric, wherein the presence of target viral RNA is Recently, the colorimetric technique, RT-LAMP assay, for the detection of viral RNA of SARS-CoV-2 using N gene-specific primers have been reported by Dao et al. [28] . The reaction setup was demonstrated to capable of detecting 100 in vitro transcription (IVT) RNA molecules. They used 1 μl of isolated RNA solution (~100 RNA molecules) in 12.5 μL reaction volume, which displayed a change of color from red to yellow following an incubation period of 30 min at 65 °C.

These results were plotted against the C t readings of RT-qPCR performed using a primer set specific for the viral E gene. For C t readings up to 30, a robust color change was noticed which determines the detection limit of this RT-LAMP assay. For the RNA samples isolated from a total of 768 pharyngeal swabs tested for COVID-19, this RT-LAMP assay detects viral RNA with the sensitivity and selectivity of 97.5% and 99.7%, respectively, for samples with C t <30. Further, the positive result of the LAMP assay was validated using the multiplexed LAMP-sequencing through the amplification of viral RNA sequences. Without RNA isolation step, direct swab (raw swab specimen) or Hot-swab (heat-treated swabs-95 °C for 5 min) to RT-LAMP assay displayed higher sensitivity for detecting samples with C t values of up to 25 and <30, respectively (Table 1) Additional techniques such as tandem MS [32, 33] , lateral flow immunoassay [34] [35] [36] , and few other enzymatic assays are established for the diagnosis of COVID-19 [37] [38] [39] [40] [41] . Tandem MS is a sophisticated technique operated by highly skilled experts and it requires relatively more time for sample preparation, data acquisition, and processing. Due to the lack of highly skilled personnel, it is not logical to implement Tandem MS for the rapid detection of viral infections during pandemic situations; however, in terms of analysis of samples for exploratory research, this method stands highly important.

As a natural response to viral infections, antibodies are secreted by the immune system through a regulated host defense mechanism. Antibodies profiling during the early phase of COVID-19 infection, as a part of serological testing, stands important [42] . In another study, 

In conclusion, an emphasis on the use of well-established spectral techniques for the swift disclosure of SARS-CoV-2 has been presented. The use of MS and colorimetric detection system, RT-LAMP assay using UV/Vis spectroscopy were developed to meet the limitations of RT-PCR used for the diagnosis of COVID-19. From a diagnostic point of view, the advantages of these methods were the overall reduced false-negative rate, lesser time of detection, low sample requirement, and improved LOD of viral RNA. Importantly, these methods can detect the positive infected cases of COVID-19 directly using nasal swab specimens without the requirement of the RNA isolation step. Collectively, these features would be useful to diagnose the COVID-19 at a faster rate, and eventually, the spread of viral infections during the current pandemic can be controlled to a great extent. Soon, these techniques may be extended to study other infectious diseases. Indeed! the MALDI-TOF technology has been already demonstrated to recognize an ample mix of viruses such as human herpesviruses (HHV) [48] , influenza viruses [49] , and for the diseases that are ascribed to intense enterovirus infections such as echovirus, coxsackievirus A and B, and poliovirus [50, 51] . Further studies on the improvement of LAMP primers design could enhance the sensitivity and specificity of the RT-LAMP assays. Additionally, the development of a cost-effective, single-step, and closed-tube isothermal PCR assay may allow the assay to conduct a largescale diagnosis of the COVID-19 virus with a very high detection capacity. Future progress in the construction of more handy POC devices consists of portable power sources and their inclusion with the RT-LAMP tools could further simplify the assay operability. Finally, all these aforementioned aspects of both the MALDI-MS and RT-LAMP assay may lead to building a potential strategy for rapidly defending the ongoing global health pandemic.

Proc. Natl. Acad. Sci

Rapid and Sensitive Detection of anti-SARS-CoV-2 IgG

SK and AAV conceptualized the review outline. AM, AF, SK performed literature survey, compared diagnostic techniques and wrote the article. AAV supervised the article preparation

We thank the Department of Science and Technology (DST), India, and Council of Scientific and 

The authors declare no competing financial interests and no conflicts of interest.

☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.☐The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:No financial interest