Limits...
Evaluation of Signature Erosion in Ebola Virus Due to Genomic Drift and Its Impact on the Performance of Diagnostic Assays.

Sozhamannan S, Holland MY, Hall AT, Negrón DA, Ivancich M, Koehler JW, Minogue TD, Campbell CE, Berger WJ, Christopher GW, Goodwin BG, Smith MA - Viruses (2015)

Bottom Line: Genome sequence analyses of the 2014 Ebola Virus (EBOV) isolates revealed a potential problem with the diagnostic assays currently in use; i.e., drifting genomic profiles of the virus may affect the sensitivity or even produce false-negative results.We evaluated signature erosion in ebolavirus molecular assays using an in silico approach and found frequent potential false-negative and false-positive results.The new signatures identified in this study may be evaluated with real-time reverse transcription PCR (rRT-PCR) assay development and validation.

View Article: PubMed Central - PubMed

Affiliation: Critical Reagents Program, Medical Countermeasure Systems Annex, 110 Thomas Johnson Drive, Frederick, MD 21702, USA. shanmuga.sozhamannan.ctr@mail.mil.

ABSTRACT
Genome sequence analyses of the 2014 Ebola Virus (EBOV) isolates revealed a potential problem with the diagnostic assays currently in use; i.e., drifting genomic profiles of the virus may affect the sensitivity or even produce false-negative results. We evaluated signature erosion in ebolavirus molecular assays using an in silico approach and found frequent potential false-negative and false-positive results. We further empirically evaluated many EBOV assays, under real time PCR conditions using EBOV Kikwit (1995) and Makona (2014) RNA templates. These results revealed differences in performance between assays but were comparable between the old and new EBOV templates. Using a whole genome approach and a novel algorithm, termed BioVelocity, we identified new signatures that are unique to each of EBOV, Sudan virus (SUDV), and Reston virus (RESTV). Interestingly, many of the current assay signatures do not fall within these regions, indicating a potential drawback in the past assay design strategies. The new signatures identified in this study may be evaluated with real-time reverse transcription PCR (rRT-PCR) assay development and validation. In addition, we discuss regulatory implications and timely availability to impact a rapidly evolving outbreak using existing but perhaps less than optimal assays versus redesign these assays for addressing genomic changes.

No MeSH data available.


Related in: MedlinePlus

Heat map of assay and amplicon hits for each assay based on percentage mismatches between the reference and various GenBank sequences. The various assays, numbered in the same order as in Table S1, are grouped according to specificity of the assay; i.e., EBOV (1–17), BDBV (19–20), TAFV (21–22), RESTV (23–26), and SUDV (27–30). The GenBank accession numbers are provided on the left, and the corresponding ebolaviruses and isolation countries are labeled on the right. The country of isolation is based on the information provided in GenBank entries. Country codes: AGO-Angola; CIV-Ivory Coast; COD-Democratic Republic of Congo; COG-Congo; GAB-Gabon; GBR-United Kingdom; GIN-Guinea; LBR-Liberia; PHL-Philippines; RUS-Russian Federation; SLE-Sierra Leone; SUD-Sudan; UGA-Uganda; UNK-Unknown; USA-United States of America; ZAR-South Africa. Assays/amplicon hits that match at 100/100 rule are in green, and others with various percent mismatches are color coded according to the scale at the bottom. (A) represents the sensitivity (color) and specificity (species) of true-positive hits to the respective assays; (B) represents the sensitivity (color) and lack of specificity (cross reactivity) to the respective assays. False-positive (#10, #16), true-negatives (#4, #24, #27). Note that the heat index scale is different in panels A and B.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4488730&req=5

viruses-07-02763-f002: Heat map of assay and amplicon hits for each assay based on percentage mismatches between the reference and various GenBank sequences. The various assays, numbered in the same order as in Table S1, are grouped according to specificity of the assay; i.e., EBOV (1–17), BDBV (19–20), TAFV (21–22), RESTV (23–26), and SUDV (27–30). The GenBank accession numbers are provided on the left, and the corresponding ebolaviruses and isolation countries are labeled on the right. The country of isolation is based on the information provided in GenBank entries. Country codes: AGO-Angola; CIV-Ivory Coast; COD-Democratic Republic of Congo; COG-Congo; GAB-Gabon; GBR-United Kingdom; GIN-Guinea; LBR-Liberia; PHL-Philippines; RUS-Russian Federation; SLE-Sierra Leone; SUD-Sudan; UGA-Uganda; UNK-Unknown; USA-United States of America; ZAR-South Africa. Assays/amplicon hits that match at 100/100 rule are in green, and others with various percent mismatches are color coded according to the scale at the bottom. (A) represents the sensitivity (color) and specificity (species) of true-positive hits to the respective assays; (B) represents the sensitivity (color) and lack of specificity (cross reactivity) to the respective assays. False-positive (#10, #16), true-negatives (#4, #24, #27). Note that the heat index scale is different in panels A and B.

Mentions: In order to more quantitatively define the sensitivity and specificity of these assays against every GenBank match, a heat map was generated [23]. In this case, sensitivity for each assay is represented in the heat map as the percentage of mismatches in individual GenBank sequences against signatures (primers/probe/amplicon) (Figure 2a representing true-positives), and the lack of specificity of the assays is represented by the reactivity of non-target organisms for any given species-specific assay (Figure 2b). The primary data used to generate the heat map is presented in Supplementary Table S2. The sensitivity level is indicated by the color; i.e., the greenest features represent perfect assays with 0 mismatches with the shift towards red being not optimal or poor assays containing mismatches (Figure 2a,b) which varied with different assays. In general, species-specific assays reacted to the appropriate species sequences with a few exceptions (see below). The assays that cross-react to other species are presented in Figure 2b. The assays that did not produce an assay hit but an amplicon hit (true-negatives) (EboZNP (#4), Ebola MGB-RESTV (#24) and Ebola Sudan-MGB (#27) and those assays that produced an assay hit but did not produce an amplicon hit because of a failure to pass the 85/90 threshold are represented in panel B of the heat map; (EBO-GP (#10) against RESTV and FiloAB (#16) against MARV and SUDV) (Figure 2b). In the latter case, EBO-GP produced false-positive results while FiloAB hits were marked as true-positives since this pan assay is intended to detect all filoviruses.


Evaluation of Signature Erosion in Ebola Virus Due to Genomic Drift and Its Impact on the Performance of Diagnostic Assays.

Sozhamannan S, Holland MY, Hall AT, Negrón DA, Ivancich M, Koehler JW, Minogue TD, Campbell CE, Berger WJ, Christopher GW, Goodwin BG, Smith MA - Viruses (2015)

Heat map of assay and amplicon hits for each assay based on percentage mismatches between the reference and various GenBank sequences. The various assays, numbered in the same order as in Table S1, are grouped according to specificity of the assay; i.e., EBOV (1–17), BDBV (19–20), TAFV (21–22), RESTV (23–26), and SUDV (27–30). The GenBank accession numbers are provided on the left, and the corresponding ebolaviruses and isolation countries are labeled on the right. The country of isolation is based on the information provided in GenBank entries. Country codes: AGO-Angola; CIV-Ivory Coast; COD-Democratic Republic of Congo; COG-Congo; GAB-Gabon; GBR-United Kingdom; GIN-Guinea; LBR-Liberia; PHL-Philippines; RUS-Russian Federation; SLE-Sierra Leone; SUD-Sudan; UGA-Uganda; UNK-Unknown; USA-United States of America; ZAR-South Africa. Assays/amplicon hits that match at 100/100 rule are in green, and others with various percent mismatches are color coded according to the scale at the bottom. (A) represents the sensitivity (color) and specificity (species) of true-positive hits to the respective assays; (B) represents the sensitivity (color) and lack of specificity (cross reactivity) to the respective assays. False-positive (#10, #16), true-negatives (#4, #24, #27). Note that the heat index scale is different in panels A and B.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4488730&req=5

viruses-07-02763-f002: Heat map of assay and amplicon hits for each assay based on percentage mismatches between the reference and various GenBank sequences. The various assays, numbered in the same order as in Table S1, are grouped according to specificity of the assay; i.e., EBOV (1–17), BDBV (19–20), TAFV (21–22), RESTV (23–26), and SUDV (27–30). The GenBank accession numbers are provided on the left, and the corresponding ebolaviruses and isolation countries are labeled on the right. The country of isolation is based on the information provided in GenBank entries. Country codes: AGO-Angola; CIV-Ivory Coast; COD-Democratic Republic of Congo; COG-Congo; GAB-Gabon; GBR-United Kingdom; GIN-Guinea; LBR-Liberia; PHL-Philippines; RUS-Russian Federation; SLE-Sierra Leone; SUD-Sudan; UGA-Uganda; UNK-Unknown; USA-United States of America; ZAR-South Africa. Assays/amplicon hits that match at 100/100 rule are in green, and others with various percent mismatches are color coded according to the scale at the bottom. (A) represents the sensitivity (color) and specificity (species) of true-positive hits to the respective assays; (B) represents the sensitivity (color) and lack of specificity (cross reactivity) to the respective assays. False-positive (#10, #16), true-negatives (#4, #24, #27). Note that the heat index scale is different in panels A and B.
Mentions: In order to more quantitatively define the sensitivity and specificity of these assays against every GenBank match, a heat map was generated [23]. In this case, sensitivity for each assay is represented in the heat map as the percentage of mismatches in individual GenBank sequences against signatures (primers/probe/amplicon) (Figure 2a representing true-positives), and the lack of specificity of the assays is represented by the reactivity of non-target organisms for any given species-specific assay (Figure 2b). The primary data used to generate the heat map is presented in Supplementary Table S2. The sensitivity level is indicated by the color; i.e., the greenest features represent perfect assays with 0 mismatches with the shift towards red being not optimal or poor assays containing mismatches (Figure 2a,b) which varied with different assays. In general, species-specific assays reacted to the appropriate species sequences with a few exceptions (see below). The assays that cross-react to other species are presented in Figure 2b. The assays that did not produce an assay hit but an amplicon hit (true-negatives) (EboZNP (#4), Ebola MGB-RESTV (#24) and Ebola Sudan-MGB (#27) and those assays that produced an assay hit but did not produce an amplicon hit because of a failure to pass the 85/90 threshold are represented in panel B of the heat map; (EBO-GP (#10) against RESTV and FiloAB (#16) against MARV and SUDV) (Figure 2b). In the latter case, EBO-GP produced false-positive results while FiloAB hits were marked as true-positives since this pan assay is intended to detect all filoviruses.

Bottom Line: Genome sequence analyses of the 2014 Ebola Virus (EBOV) isolates revealed a potential problem with the diagnostic assays currently in use; i.e., drifting genomic profiles of the virus may affect the sensitivity or even produce false-negative results.We evaluated signature erosion in ebolavirus molecular assays using an in silico approach and found frequent potential false-negative and false-positive results.The new signatures identified in this study may be evaluated with real-time reverse transcription PCR (rRT-PCR) assay development and validation.

View Article: PubMed Central - PubMed

Affiliation: Critical Reagents Program, Medical Countermeasure Systems Annex, 110 Thomas Johnson Drive, Frederick, MD 21702, USA. shanmuga.sozhamannan.ctr@mail.mil.

ABSTRACT
Genome sequence analyses of the 2014 Ebola Virus (EBOV) isolates revealed a potential problem with the diagnostic assays currently in use; i.e., drifting genomic profiles of the virus may affect the sensitivity or even produce false-negative results. We evaluated signature erosion in ebolavirus molecular assays using an in silico approach and found frequent potential false-negative and false-positive results. We further empirically evaluated many EBOV assays, under real time PCR conditions using EBOV Kikwit (1995) and Makona (2014) RNA templates. These results revealed differences in performance between assays but were comparable between the old and new EBOV templates. Using a whole genome approach and a novel algorithm, termed BioVelocity, we identified new signatures that are unique to each of EBOV, Sudan virus (SUDV), and Reston virus (RESTV). Interestingly, many of the current assay signatures do not fall within these regions, indicating a potential drawback in the past assay design strategies. The new signatures identified in this study may be evaluated with real-time reverse transcription PCR (rRT-PCR) assay development and validation. In addition, we discuss regulatory implications and timely availability to impact a rapidly evolving outbreak using existing but perhaps less than optimal assays versus redesign these assays for addressing genomic changes.

No MeSH data available.


Related in: MedlinePlus