Deep Impact of Random Amplification and Library Construction Methods on Viral Metagenomics Results

Abstract

Clinical metagenomics is a broad-range agnostic detection method of pathogens, including novel microorganisms. A major limit is the low pathogen load compared to the high background of host nucleic acids. To overcome this issue, several solutions exist, such as applying a very high depth of sequencing, or performing a relative enrichment of viral genomes associated with capsids. At the end, the quantity of total nucleic acids is often below the concentrations recommended by the manufacturers of library kits, which necessitates to random amplify nucleic acids. Using a pool of 26 viruses representative of viral diversity, we observed a deep impact of the nature of sample (total nucleic acids versus RNA only), the reverse transcription, the random amplification and library construction method on virus recovery. We further optimized the two most promising methods and assessed their performance with fully characterized reference virus stocks. Good genome coverage and limit of detection lower than 100 or 1000 genome copies per mL of plasma, depending on the genome viral type, were obtained from a three million reads dataset. Our study reveals that optimized random amplification is a technique of choice when insufficient amounts of nucleic acid are available for direct libraries constructions.

Keywords: random amplification; sensitivity; viral metagenomics.

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Figure 1

Experimental design with Virus Multiplex Reference Panel (VMRP). Six protocols including reverse transcription step, random pre-amplification step and subsequent library constructions were compared to a protocol without pre-amplification (NoAmp). A first experiment was carried out directly on a mix of 26 viruses representative of the diversity of viruses (VMRP) and the second experiment on VMRP diluted in a plasma matrix (ratio: 1/10). Both fractions, total nucleic acid fraction (NA) and RNA only, were used.

Figure 2

Experimental design with WHO reference virus stocks (WRVS). The experiment was designed to quantify the viral detection level of two MALBAC-based methods compared to the protocol without pre-amplification (NoAmp). The NA fraction of WRVS (mix of quantified reference virus stocks) was used at a final concentration of 104, 103 or 102 viral gc/mL of plasma matrix.

Figure 3

Proportion of spiked viruses (upper panels (A,C)) and viral genomic groups (lower panels (B,D)) in the VMRP based on the weighted contigs and singletons (WNCS). (A,B) In the NA fraction; (C,D) in the RNA fraction. Comparison of viral genomic groups is shown for target viruses (top) and for all viruses (bottom).

Figure 3

Proportion of spiked viruses (upper panels (A,C)) and viral genomic groups (lower panels (B,D)) in the VMRP based on the weighted contigs and singletons (WNCS). (A,B) In the NA fraction; (C,D) in the RNA fraction. Comparison of viral genomic groups is shown for target viruses (top) and for all viruses (bottom).

Figure 4

Heatmap of whole genome coverage of viruses of VMRP from seven methods in NA and RNA fractions. The genome fraction for each virus is in row and methods for both NA and RNA fractions are in column. Viruses are color-coded according to their genomic group (dsDNA, ssDNA, dsRNA, ss(+) RNA and ss(-) RNA). An asterisk * indicates the non-enveloped viruses. Norovirus GI detected by no method was removed. Ranges of Ct value previously determined [31] were reported per viral cluster.

Figure 5

Combined results of virus detection genome fraction of viruses detected in crude VMRP and in VMRP spiked in plasma sample with seven methods and viral detection in VMRP spiked in plasma sample. (A) in NA fraction; (B) in RNA fraction. Stacked histogram represents the genome fraction (%) in crude VMRP (blue) and in spiked plasma matrix (orange). DOPlify1 and DOPlify2 referred at different number of amplification cycles. Note that MATQ was not assessed in plasma sample.

Figure 5

Combined results of virus detection genome fraction of viruses detected in crude VMRP and in VMRP spiked in plasma sample with seven methods and viral detection in VMRP spiked in plasma sample. (A) in NA fraction; (B) in RNA fraction. Stacked histogram represents the genome fraction (%) in crude VMRP (blue) and in spiked plasma matrix (orange). DOPlify1 and DOPlify2 referred at different number of amplification cycles. Note that MATQ was not assessed in plasma sample.

Figure 6

Effect of denaturation temperature of dsRNA virus prior to reverse-transcription on the treated sample with or without benzonase. Each viral stock (HHV-4, REO L1, RSV, FeLV) was diluted to 5 × 106 genome-copies per mL for the NA isolation.

Figure 7

Genome coverage of reference virus stocks (HHV-4, PCV1, REO1, HRSV, FeLV, plus SMRV) spiked in plasma sample pool at 102, 103 or 104 genome-copies per mL (gc/mL) for NoAmp, MALBAC and MALBAC_V2 methods.