| Server/URL . | Functionalitya . | Operating principlesb . | Performancec . | Inputsd . | Outputs . | Advantages . | Limitations . | Year . |
|---|---|---|---|---|---|---|---|---|
| Predict DRMs from insect sequence | ||||||||
| ACE http://genome.zju.edu.cn/software/ace/ | Detect insecticide resistance mutations in AchE by RNA-Seq data | BWT-based sequence mapping | – | FASTA or FASTQ | Mutation frequency, Resistance frequency | The first tool to detect DRMs from RNA-Seq data, can detect resistant reads at low frequency | Only one target resistance mutation can be detected currently | 2017 |
| FastD http://www.insect-genome.com/fastd | Detect insecticide resistance target-site mutations by RNA-Seq data | BWT–based sequence mapping | AUC: 0.87, R2 = 0.834, AC: 89.7% | cDNA sequences, SAM file | Mutation frequency, Resistance frequency | Can identify the new target-site mutations, using SAM files as input which can analyze the samples more quickly | The accuracy of mutation frequency is limited by the fact that RNA-Seq reads from pooled sample have potentially different levels of contribution from each insect sample and allele | 2019 |
| Predict DRMs from microorganism sequence | ||||||||
| LRE-Finder https://cge.food.dtu.dk/services/LRE-Finder-1.0/ | Detects the 23S rRNA mutations and linezolid resistance in enterococci by WGS data | KMA–based sequence mapping | AC: 100% | Elm database, threshholds, FASTA or FASTQ | Mutations, wild-type ratio, MT type ratio and predicted phenotype | The first report of a G2505A mutation detected in vivo in an E. faecium isolate from a patient | Using draft as sembly sequences will fail to detect mutations in 23S, when these mutations are constituting only a minority of the bases in the given position | 2019 |
| PointFinder https://cge.cbs.dtu.dk/services/ | Detects AMR chromosomal point mutations in bacteria | BLAST-based sequence alignment | AC: 98.4% | FASTQ | – | The output from the web tool is easily understandable | Low accessibility | 2018 |
| MinVar http://git.io/minvar | Detects minority variants in HIV-1 and HCV populations | BWA (BWT-based) sequence mapping | – | FASTQ | A table with amino acid mutations with respect to HIV-1 consensus B, annotated according to the class of resistance defined in the Stanford HIVdb | Detect DRMs without the need to perform additional bioinformatics analysis; Be compatible with a diverse range of sequencing platforms | There is no check for minimum acceptable and uniform coverage. For anomalous samples, a strategy to correct this skew is not chosen | 2017 |
| GWAMAR http://bioputer.mimuw.edu.pl/gwamar/ | Detects DRMs in bacteria from WGS data | MSA, TGH | AUC: 0.28, 0.43 | Mutations, drug resistance profiles, phylogenetic tree | Scored list of putative associations of drug resistance with mutations | Designed a new statistical score TGH | (i) it doesn’t consider or predict epistatic interactions between mutations. (ii) it considers only genomic changes ignoring levels of gene expression. (iii) it provides putative in silico associations which should be subjected to further investigation in wet lab experiments. | 2014 |
| HIVfird www.hivfird.ics.ufba.br | Detects mutatons in HIV-1 sequences that confer resistance to Enfuvirtide | Kalign-based sequence alignment | – | DNA FASTA | HTML file return from server with detection report | The first software to predict the resistance of HIV-1 strains to the fusion inhibitors based on the virus DNA sequence | Only nucleotide sequences can be used as input, protein sequences cannot be used as input | 2019 |
| Resistance Sniffer http://resistance-sniffer.bi.up.ac.za/ | Predicts drug resistance patterns of MTB isolates | BWT-based sequence mapping | – | FASTA/FASTQ | A bar plot of the probability that the strain is drug sensitive or drug resistant to the 13 antibiotics | Can be used at different stages of whole genome completion | Predictable anti-TB drugs are limited | 2019 |
| Mykrobe predictor https://www.mykrobe.com/ | Predicts drug resistance for MTB and SA from WGS data | BWT-based sequence mapping | SE/SP: 99.1%/99.6%; 82.6%/98.5% | FASTQ | Clinician-friendly report | A system robust to mixture | Batch uploads are not allowed, can’t interpret low frequency mutations with some of the platforms completely insensitive to indels and variants in promoter regions | 2015 |
| TB-Profiler https://tbdr.lshtm.ac.uk/ | Detects anti-TB drug resistance from WGS data | BWA (BWT-based) sequence alignment | – | FASTQ | HTML with drug resistance profile/lineages | The mutation library is more accurate than current commercial molecular tests and alternative mutation databases | Batch uploads are not allowed, can’t interpret low frequency mutations with some of the platforms completely insensitive to indels and variants in promoter regions | 2015, 2019 |
| PhyResSE http://phyresse.org | Delineates drug resistance of MTB from WGS data | BLAST-based sequence mapping | AC: 97.83%–100% | FASTQ | HTML with drug resistance profile and lineages | Simple to use, befits human diagnostics | Can’t interpret low frequency mutations with some of the platforms completely insensitive to indels and variants in promoter regions | 2015 |
| KvarQ http://www.swisstph.ch/kvarq. | Detects DRMs in bacterial from WGS data | BWA (BWT-based) sequence alignment | AC: >99% | FASTQ | A text file in JavaScript Object Notation format | Directly extracts relevant information from fastq files, easy to use | Can’t interpret low frequency mutations with some of the platforms completely insensitive to indels and variants in promoter regions | 2014 |
| CASTB http://castb.ri.ncgm.go.jp/CASTB | Predicts drug resistance for MTB from WGS data | – | – | FASTA/ FASTQ | Spoligotypes, VNTR, LSP lineages and SNP based tree with e-mail notification | CASTB is a useful tool for identifying strains from WGS data, even when bioinformatics knowledge is limited. | Batch uploads are not allowed,can’t interpret low frequency mutations with some of the platforms completely insensitive to indels and variants in promoter regions | 2015 |
| GenTB https://gentb.hms.harvard.edu | For analyzing and predicting drug resistances to MTB | MEM–Align–based sequence alignment | SE/SP: GenTB-RF: 77.6%, 96.2% GenTB-WDNN: 75.4%, 96.1% | FASTQ files and varient call file | Mutation frequency | Users can choose between two potential predictors, a RF classifier and a Wide and Deep Neural Network | Need to quality control input sequence data before prediction; multipoint mutations cannot be predicted | 2021 |
| AMRFinderPlus https://www.ncbi.nlm.nih.gov/pathogens/antimicrobial resistance/AMRFinder/ | Predicts drug resistance-associated point mutations | BLAST-based sequence alignment | – | FASTA | Report | Can detect acquired genes and point mutations in both protein and nucleotide sequence | Not easy to use | 2021 |
| SAM-TB https://samtb.uni-medica.com/ | Detects MTB drug resistance and transmission | BWA (BWT-based) sequence mapping | SE: 93.9%, SP: 96.2% | FASTQ | Mutation frequency, mutation details | Integrates drug-resistance prediction with strain genetic relationships and species identification of nontuberculous mycobacteria | Predictable anti-TB drugs are limited | 2022 |
| Server/URL . | Functionalitya . | Operating principlesb . | Performancec . | Inputsd . | Outputs . | Advantages . | Limitations . | Year . |
|---|---|---|---|---|---|---|---|---|
| Predict DRMs from insect sequence | ||||||||
| ACE http://genome.zju.edu.cn/software/ace/ | Detect insecticide resistance mutations in AchE by RNA-Seq data | BWT-based sequence mapping | – | FASTA or FASTQ | Mutation frequency, Resistance frequency | The first tool to detect DRMs from RNA-Seq data, can detect resistant reads at low frequency | Only one target resistance mutation can be detected currently | 2017 |
| FastD http://www.insect-genome.com/fastd | Detect insecticide resistance target-site mutations by RNA-Seq data | BWT–based sequence mapping | AUC: 0.87, R2 = 0.834, AC: 89.7% | cDNA sequences, SAM file | Mutation frequency, Resistance frequency | Can identify the new target-site mutations, using SAM files as input which can analyze the samples more quickly | The accuracy of mutation frequency is limited by the fact that RNA-Seq reads from pooled sample have potentially different levels of contribution from each insect sample and allele | 2019 |
| Predict DRMs from microorganism sequence | ||||||||
| LRE-Finder https://cge.food.dtu.dk/services/LRE-Finder-1.0/ | Detects the 23S rRNA mutations and linezolid resistance in enterococci by WGS data | KMA–based sequence mapping | AC: 100% | Elm database, threshholds, FASTA or FASTQ | Mutations, wild-type ratio, MT type ratio and predicted phenotype | The first report of a G2505A mutation detected in vivo in an E. faecium isolate from a patient | Using draft as sembly sequences will fail to detect mutations in 23S, when these mutations are constituting only a minority of the bases in the given position | 2019 |
| PointFinder https://cge.cbs.dtu.dk/services/ | Detects AMR chromosomal point mutations in bacteria | BLAST-based sequence alignment | AC: 98.4% | FASTQ | – | The output from the web tool is easily understandable | Low accessibility | 2018 |
| MinVar http://git.io/minvar | Detects minority variants in HIV-1 and HCV populations | BWA (BWT-based) sequence mapping | – | FASTQ | A table with amino acid mutations with respect to HIV-1 consensus B, annotated according to the class of resistance defined in the Stanford HIVdb | Detect DRMs without the need to perform additional bioinformatics analysis; Be compatible with a diverse range of sequencing platforms | There is no check for minimum acceptable and uniform coverage. For anomalous samples, a strategy to correct this skew is not chosen | 2017 |
| GWAMAR http://bioputer.mimuw.edu.pl/gwamar/ | Detects DRMs in bacteria from WGS data | MSA, TGH | AUC: 0.28, 0.43 | Mutations, drug resistance profiles, phylogenetic tree | Scored list of putative associations of drug resistance with mutations | Designed a new statistical score TGH | (i) it doesn’t consider or predict epistatic interactions between mutations. (ii) it considers only genomic changes ignoring levels of gene expression. (iii) it provides putative in silico associations which should be subjected to further investigation in wet lab experiments. | 2014 |
| HIVfird www.hivfird.ics.ufba.br | Detects mutatons in HIV-1 sequences that confer resistance to Enfuvirtide | Kalign-based sequence alignment | – | DNA FASTA | HTML file return from server with detection report | The first software to predict the resistance of HIV-1 strains to the fusion inhibitors based on the virus DNA sequence | Only nucleotide sequences can be used as input, protein sequences cannot be used as input | 2019 |
| Resistance Sniffer http://resistance-sniffer.bi.up.ac.za/ | Predicts drug resistance patterns of MTB isolates | BWT-based sequence mapping | – | FASTA/FASTQ | A bar plot of the probability that the strain is drug sensitive or drug resistant to the 13 antibiotics | Can be used at different stages of whole genome completion | Predictable anti-TB drugs are limited | 2019 |
| Mykrobe predictor https://www.mykrobe.com/ | Predicts drug resistance for MTB and SA from WGS data | BWT-based sequence mapping | SE/SP: 99.1%/99.6%; 82.6%/98.5% | FASTQ | Clinician-friendly report | A system robust to mixture | Batch uploads are not allowed, can’t interpret low frequency mutations with some of the platforms completely insensitive to indels and variants in promoter regions | 2015 |
| TB-Profiler https://tbdr.lshtm.ac.uk/ | Detects anti-TB drug resistance from WGS data | BWA (BWT-based) sequence alignment | – | FASTQ | HTML with drug resistance profile/lineages | The mutation library is more accurate than current commercial molecular tests and alternative mutation databases | Batch uploads are not allowed, can’t interpret low frequency mutations with some of the platforms completely insensitive to indels and variants in promoter regions | 2015, 2019 |
| PhyResSE http://phyresse.org | Delineates drug resistance of MTB from WGS data | BLAST-based sequence mapping | AC: 97.83%–100% | FASTQ | HTML with drug resistance profile and lineages | Simple to use, befits human diagnostics | Can’t interpret low frequency mutations with some of the platforms completely insensitive to indels and variants in promoter regions | 2015 |
| KvarQ http://www.swisstph.ch/kvarq. | Detects DRMs in bacterial from WGS data | BWA (BWT-based) sequence alignment | AC: >99% | FASTQ | A text file in JavaScript Object Notation format | Directly extracts relevant information from fastq files, easy to use | Can’t interpret low frequency mutations with some of the platforms completely insensitive to indels and variants in promoter regions | 2014 |
| CASTB http://castb.ri.ncgm.go.jp/CASTB | Predicts drug resistance for MTB from WGS data | – | – | FASTA/ FASTQ | Spoligotypes, VNTR, LSP lineages and SNP based tree with e-mail notification | CASTB is a useful tool for identifying strains from WGS data, even when bioinformatics knowledge is limited. | Batch uploads are not allowed,can’t interpret low frequency mutations with some of the platforms completely insensitive to indels and variants in promoter regions | 2015 |
| GenTB https://gentb.hms.harvard.edu | For analyzing and predicting drug resistances to MTB | MEM–Align–based sequence alignment | SE/SP: GenTB-RF: 77.6%, 96.2% GenTB-WDNN: 75.4%, 96.1% | FASTQ files and varient call file | Mutation frequency | Users can choose between two potential predictors, a RF classifier and a Wide and Deep Neural Network | Need to quality control input sequence data before prediction; multipoint mutations cannot be predicted | 2021 |
| AMRFinderPlus https://www.ncbi.nlm.nih.gov/pathogens/antimicrobial resistance/AMRFinder/ | Predicts drug resistance-associated point mutations | BLAST-based sequence alignment | – | FASTA | Report | Can detect acquired genes and point mutations in both protein and nucleotide sequence | Not easy to use | 2021 |
| SAM-TB https://samtb.uni-medica.com/ | Detects MTB drug resistance and transmission | BWA (BWT-based) sequence mapping | SE: 93.9%, SP: 96.2% | FASTQ | Mutation frequency, mutation details | Integrates drug-resistance prediction with strain genetic relationships and species identification of nontuberculous mycobacteria | Predictable anti-TB drugs are limited | 2022 |
aAbbreviation: AchE: Acetylcholine esterase; WGS: Whole Genome Sequencing; AMR: Antimicrobial resistance; DRMs: Drug resistance mutations; MTB: M. tuberculosis; SA: S. aureus.
bAbbreviation: BWT: Burrows–Wheeler Transform, KMA: K-mer alignment, uses k-mer seeding to speed up mapping and the Needleman–Wunsch algorithm to accurately align extensions from k-mer seeds. BWA: Burrows-Wheeler Alignment, a short read alignment with BWT. MSA: multiple sequence alignment. TGH: A new statistical score, viz tree-generalized hypergeometric score. Kalign: An MSA program that uses a SIMD (single instruction, multiple data) accelerated version of the bit-parallel Gene Myers algorithm. MEM-Align: A fast semi-global alignment algorithm for short DNA sequences that allows for affine-gap scoring and exploit sequence similarity. BLAST: The Basic Local Alignment Search Tool.
cPerformance: The sample information of the performance corresponding to these severs is provided in detail. FastD: They detected 469 (89.7%) variants among the inserted variants, calling performance using AUC in ROC curve. ROC with an AUC of 0.870 indicated a reliable calling performance. They compared the detected allele frequencies of detected variants with their set allele frequencies and found that the allele frequencies calculated by FastD-TR were highly correlated with their ‘true’ allele frequencies (R2 = 0.834; ρ < 10−16). LRE-Finder: Fastq files from 21 LRE isolates were submitted to LRE-Finder. As negative controls, fastq files from 1473 non-LRE isolates were submitted to LRE-Finder. The MICs of linezolid were determined for the 21 LRE isolates. As LRE-negative controls, 26 VRE isolates were additionally selected for linezolid MIC determination. It was validated and showed 100% concordance with phenotypic susceptibility testing. PointFinder: A total of 685 different phenotypic tests associated with chromosomal resistance to quinolones, polymyxin, rifampicin, macrolides and tetracyclines resulted in 98.4% concordance. GWAMAR: Precision-recall curves for comparison of different association scores implemented in GWAMAR. One presents results for the mtu173 dataset (39 positives; 1450 negatives), AUC = 0.28; the other for the mtu_broad dataset (75 positives; 870 negatives), AUC = 0.43. Mykrobe predictor: With SE/SP of 99.1%/99.6% across 12 antibiotics (using an independent validation set, n = 470). For MTB, the method predicts resistance with SE/SP of 82.6%/98.5% (independent validation set, n = 1609). PhyResSE: PhyResSE was tested with 92 strains from a well-characterized strain collection from Sierra Leone that comprised 44 phenotypically susceptible strains and 48 strains. 100% concordance for resistance SNPs in katG, inhA, ahpC, rrs, rpsL, embA and embC; 98.91% concordance for those in gidB and pncA; and 97.83% concordance for those in rpoB and embB. KvarQ: KvarQ successfully detect all main DRMs and phylogenetic markers in 880 bacterial whole genome sequences. The variant calls of a subset of these genomes were validated with a standard bioinformatics pipeline and revealed >99% congruency. GenTB: using a ground truth dataset of 20,408 isolates with laboratory-based drug susceptibility data. The mean sensitivities for GenTB RF and GenTB-WDNN across the nine shared drugs were 77.6% and 75.4%, respectively. The specificity: GenTB-WDNN 96.2%, and GenTB-RF 96.1%. SAM-TB: The accuracy of SAM-TB in predicting drug-resistance was assessed using 3177 sequenced clinical isolates with results of phenotypic drug-susceptibility tests (pDST). Compared to pDST, the sensitivity of SAM-TB for detecting multidrug-resistant tuberculosis was 93.9% with specificity of 96.2%. Abbreviation: AUC: Area Under Curve. AC: Accuracy. SE: Sensitivity. SP: Specificity.
dSAM file: the file of SAM format; NGS: next generation sequencing.
| Server/URL . | Functionalitya . | Operating principlesb . | Performancec . | Inputsd . | Outputs . | Advantages . | Limitations . | Year . |
|---|---|---|---|---|---|---|---|---|
| Predict DRMs from insect sequence | ||||||||
| ACE http://genome.zju.edu.cn/software/ace/ | Detect insecticide resistance mutations in AchE by RNA-Seq data | BWT-based sequence mapping | – | FASTA or FASTQ | Mutation frequency, Resistance frequency | The first tool to detect DRMs from RNA-Seq data, can detect resistant reads at low frequency | Only one target resistance mutation can be detected currently | 2017 |
| FastD http://www.insect-genome.com/fastd | Detect insecticide resistance target-site mutations by RNA-Seq data | BWT–based sequence mapping | AUC: 0.87, R2 = 0.834, AC: 89.7% | cDNA sequences, SAM file | Mutation frequency, Resistance frequency | Can identify the new target-site mutations, using SAM files as input which can analyze the samples more quickly | The accuracy of mutation frequency is limited by the fact that RNA-Seq reads from pooled sample have potentially different levels of contribution from each insect sample and allele | 2019 |
| Predict DRMs from microorganism sequence | ||||||||
| LRE-Finder https://cge.food.dtu.dk/services/LRE-Finder-1.0/ | Detects the 23S rRNA mutations and linezolid resistance in enterococci by WGS data | KMA–based sequence mapping | AC: 100% | Elm database, threshholds, FASTA or FASTQ | Mutations, wild-type ratio, MT type ratio and predicted phenotype | The first report of a G2505A mutation detected in vivo in an E. faecium isolate from a patient | Using draft as sembly sequences will fail to detect mutations in 23S, when these mutations are constituting only a minority of the bases in the given position | 2019 |
| PointFinder https://cge.cbs.dtu.dk/services/ | Detects AMR chromosomal point mutations in bacteria | BLAST-based sequence alignment | AC: 98.4% | FASTQ | – | The output from the web tool is easily understandable | Low accessibility | 2018 |
| MinVar http://git.io/minvar | Detects minority variants in HIV-1 and HCV populations | BWA (BWT-based) sequence mapping | – | FASTQ | A table with amino acid mutations with respect to HIV-1 consensus B, annotated according to the class of resistance defined in the Stanford HIVdb | Detect DRMs without the need to perform additional bioinformatics analysis; Be compatible with a diverse range of sequencing platforms | There is no check for minimum acceptable and uniform coverage. For anomalous samples, a strategy to correct this skew is not chosen | 2017 |
| GWAMAR http://bioputer.mimuw.edu.pl/gwamar/ | Detects DRMs in bacteria from WGS data | MSA, TGH | AUC: 0.28, 0.43 | Mutations, drug resistance profiles, phylogenetic tree | Scored list of putative associations of drug resistance with mutations | Designed a new statistical score TGH | (i) it doesn’t consider or predict epistatic interactions between mutations. (ii) it considers only genomic changes ignoring levels of gene expression. (iii) it provides putative in silico associations which should be subjected to further investigation in wet lab experiments. | 2014 |
| HIVfird www.hivfird.ics.ufba.br | Detects mutatons in HIV-1 sequences that confer resistance to Enfuvirtide | Kalign-based sequence alignment | – | DNA FASTA | HTML file return from server with detection report | The first software to predict the resistance of HIV-1 strains to the fusion inhibitors based on the virus DNA sequence | Only nucleotide sequences can be used as input, protein sequences cannot be used as input | 2019 |
| Resistance Sniffer http://resistance-sniffer.bi.up.ac.za/ | Predicts drug resistance patterns of MTB isolates | BWT-based sequence mapping | – | FASTA/FASTQ | A bar plot of the probability that the strain is drug sensitive or drug resistant to the 13 antibiotics | Can be used at different stages of whole genome completion | Predictable anti-TB drugs are limited | 2019 |
| Mykrobe predictor https://www.mykrobe.com/ | Predicts drug resistance for MTB and SA from WGS data | BWT-based sequence mapping | SE/SP: 99.1%/99.6%; 82.6%/98.5% | FASTQ | Clinician-friendly report | A system robust to mixture | Batch uploads are not allowed, can’t interpret low frequency mutations with some of the platforms completely insensitive to indels and variants in promoter regions | 2015 |
| TB-Profiler https://tbdr.lshtm.ac.uk/ | Detects anti-TB drug resistance from WGS data | BWA (BWT-based) sequence alignment | – | FASTQ | HTML with drug resistance profile/lineages | The mutation library is more accurate than current commercial molecular tests and alternative mutation databases | Batch uploads are not allowed, can’t interpret low frequency mutations with some of the platforms completely insensitive to indels and variants in promoter regions | 2015, 2019 |
| PhyResSE http://phyresse.org | Delineates drug resistance of MTB from WGS data | BLAST-based sequence mapping | AC: 97.83%–100% | FASTQ | HTML with drug resistance profile and lineages | Simple to use, befits human diagnostics | Can’t interpret low frequency mutations with some of the platforms completely insensitive to indels and variants in promoter regions | 2015 |
| KvarQ http://www.swisstph.ch/kvarq. | Detects DRMs in bacterial from WGS data | BWA (BWT-based) sequence alignment | AC: >99% | FASTQ | A text file in JavaScript Object Notation format | Directly extracts relevant information from fastq files, easy to use | Can’t interpret low frequency mutations with some of the platforms completely insensitive to indels and variants in promoter regions | 2014 |
| CASTB http://castb.ri.ncgm.go.jp/CASTB | Predicts drug resistance for MTB from WGS data | – | – | FASTA/ FASTQ | Spoligotypes, VNTR, LSP lineages and SNP based tree with e-mail notification | CASTB is a useful tool for identifying strains from WGS data, even when bioinformatics knowledge is limited. | Batch uploads are not allowed,can’t interpret low frequency mutations with some of the platforms completely insensitive to indels and variants in promoter regions | 2015 |
| GenTB https://gentb.hms.harvard.edu | For analyzing and predicting drug resistances to MTB | MEM–Align–based sequence alignment | SE/SP: GenTB-RF: 77.6%, 96.2% GenTB-WDNN: 75.4%, 96.1% | FASTQ files and varient call file | Mutation frequency | Users can choose between two potential predictors, a RF classifier and a Wide and Deep Neural Network | Need to quality control input sequence data before prediction; multipoint mutations cannot be predicted | 2021 |
| AMRFinderPlus https://www.ncbi.nlm.nih.gov/pathogens/antimicrobial resistance/AMRFinder/ | Predicts drug resistance-associated point mutations | BLAST-based sequence alignment | – | FASTA | Report | Can detect acquired genes and point mutations in both protein and nucleotide sequence | Not easy to use | 2021 |
| SAM-TB https://samtb.uni-medica.com/ | Detects MTB drug resistance and transmission | BWA (BWT-based) sequence mapping | SE: 93.9%, SP: 96.2% | FASTQ | Mutation frequency, mutation details | Integrates drug-resistance prediction with strain genetic relationships and species identification of nontuberculous mycobacteria | Predictable anti-TB drugs are limited | 2022 |
| Server/URL . | Functionalitya . | Operating principlesb . | Performancec . | Inputsd . | Outputs . | Advantages . | Limitations . | Year . |
|---|---|---|---|---|---|---|---|---|
| Predict DRMs from insect sequence | ||||||||
| ACE http://genome.zju.edu.cn/software/ace/ | Detect insecticide resistance mutations in AchE by RNA-Seq data | BWT-based sequence mapping | – | FASTA or FASTQ | Mutation frequency, Resistance frequency | The first tool to detect DRMs from RNA-Seq data, can detect resistant reads at low frequency | Only one target resistance mutation can be detected currently | 2017 |
| FastD http://www.insect-genome.com/fastd | Detect insecticide resistance target-site mutations by RNA-Seq data | BWT–based sequence mapping | AUC: 0.87, R2 = 0.834, AC: 89.7% | cDNA sequences, SAM file | Mutation frequency, Resistance frequency | Can identify the new target-site mutations, using SAM files as input which can analyze the samples more quickly | The accuracy of mutation frequency is limited by the fact that RNA-Seq reads from pooled sample have potentially different levels of contribution from each insect sample and allele | 2019 |
| Predict DRMs from microorganism sequence | ||||||||
| LRE-Finder https://cge.food.dtu.dk/services/LRE-Finder-1.0/ | Detects the 23S rRNA mutations and linezolid resistance in enterococci by WGS data | KMA–based sequence mapping | AC: 100% | Elm database, threshholds, FASTA or FASTQ | Mutations, wild-type ratio, MT type ratio and predicted phenotype | The first report of a G2505A mutation detected in vivo in an E. faecium isolate from a patient | Using draft as sembly sequences will fail to detect mutations in 23S, when these mutations are constituting only a minority of the bases in the given position | 2019 |
| PointFinder https://cge.cbs.dtu.dk/services/ | Detects AMR chromosomal point mutations in bacteria | BLAST-based sequence alignment | AC: 98.4% | FASTQ | – | The output from the web tool is easily understandable | Low accessibility | 2018 |
| MinVar http://git.io/minvar | Detects minority variants in HIV-1 and HCV populations | BWA (BWT-based) sequence mapping | – | FASTQ | A table with amino acid mutations with respect to HIV-1 consensus B, annotated according to the class of resistance defined in the Stanford HIVdb | Detect DRMs without the need to perform additional bioinformatics analysis; Be compatible with a diverse range of sequencing platforms | There is no check for minimum acceptable and uniform coverage. For anomalous samples, a strategy to correct this skew is not chosen | 2017 |
| GWAMAR http://bioputer.mimuw.edu.pl/gwamar/ | Detects DRMs in bacteria from WGS data | MSA, TGH | AUC: 0.28, 0.43 | Mutations, drug resistance profiles, phylogenetic tree | Scored list of putative associations of drug resistance with mutations | Designed a new statistical score TGH | (i) it doesn’t consider or predict epistatic interactions between mutations. (ii) it considers only genomic changes ignoring levels of gene expression. (iii) it provides putative in silico associations which should be subjected to further investigation in wet lab experiments. | 2014 |
| HIVfird www.hivfird.ics.ufba.br | Detects mutatons in HIV-1 sequences that confer resistance to Enfuvirtide | Kalign-based sequence alignment | – | DNA FASTA | HTML file return from server with detection report | The first software to predict the resistance of HIV-1 strains to the fusion inhibitors based on the virus DNA sequence | Only nucleotide sequences can be used as input, protein sequences cannot be used as input | 2019 |
| Resistance Sniffer http://resistance-sniffer.bi.up.ac.za/ | Predicts drug resistance patterns of MTB isolates | BWT-based sequence mapping | – | FASTA/FASTQ | A bar plot of the probability that the strain is drug sensitive or drug resistant to the 13 antibiotics | Can be used at different stages of whole genome completion | Predictable anti-TB drugs are limited | 2019 |
| Mykrobe predictor https://www.mykrobe.com/ | Predicts drug resistance for MTB and SA from WGS data | BWT-based sequence mapping | SE/SP: 99.1%/99.6%; 82.6%/98.5% | FASTQ | Clinician-friendly report | A system robust to mixture | Batch uploads are not allowed, can’t interpret low frequency mutations with some of the platforms completely insensitive to indels and variants in promoter regions | 2015 |
| TB-Profiler https://tbdr.lshtm.ac.uk/ | Detects anti-TB drug resistance from WGS data | BWA (BWT-based) sequence alignment | – | FASTQ | HTML with drug resistance profile/lineages | The mutation library is more accurate than current commercial molecular tests and alternative mutation databases | Batch uploads are not allowed, can’t interpret low frequency mutations with some of the platforms completely insensitive to indels and variants in promoter regions | 2015, 2019 |
| PhyResSE http://phyresse.org | Delineates drug resistance of MTB from WGS data | BLAST-based sequence mapping | AC: 97.83%–100% | FASTQ | HTML with drug resistance profile and lineages | Simple to use, befits human diagnostics | Can’t interpret low frequency mutations with some of the platforms completely insensitive to indels and variants in promoter regions | 2015 |
| KvarQ http://www.swisstph.ch/kvarq. | Detects DRMs in bacterial from WGS data | BWA (BWT-based) sequence alignment | AC: >99% | FASTQ | A text file in JavaScript Object Notation format | Directly extracts relevant information from fastq files, easy to use | Can’t interpret low frequency mutations with some of the platforms completely insensitive to indels and variants in promoter regions | 2014 |
| CASTB http://castb.ri.ncgm.go.jp/CASTB | Predicts drug resistance for MTB from WGS data | – | – | FASTA/ FASTQ | Spoligotypes, VNTR, LSP lineages and SNP based tree with e-mail notification | CASTB is a useful tool for identifying strains from WGS data, even when bioinformatics knowledge is limited. | Batch uploads are not allowed,can’t interpret low frequency mutations with some of the platforms completely insensitive to indels and variants in promoter regions | 2015 |
| GenTB https://gentb.hms.harvard.edu | For analyzing and predicting drug resistances to MTB | MEM–Align–based sequence alignment | SE/SP: GenTB-RF: 77.6%, 96.2% GenTB-WDNN: 75.4%, 96.1% | FASTQ files and varient call file | Mutation frequency | Users can choose between two potential predictors, a RF classifier and a Wide and Deep Neural Network | Need to quality control input sequence data before prediction; multipoint mutations cannot be predicted | 2021 |
| AMRFinderPlus https://www.ncbi.nlm.nih.gov/pathogens/antimicrobial resistance/AMRFinder/ | Predicts drug resistance-associated point mutations | BLAST-based sequence alignment | – | FASTA | Report | Can detect acquired genes and point mutations in both protein and nucleotide sequence | Not easy to use | 2021 |
| SAM-TB https://samtb.uni-medica.com/ | Detects MTB drug resistance and transmission | BWA (BWT-based) sequence mapping | SE: 93.9%, SP: 96.2% | FASTQ | Mutation frequency, mutation details | Integrates drug-resistance prediction with strain genetic relationships and species identification of nontuberculous mycobacteria | Predictable anti-TB drugs are limited | 2022 |
aAbbreviation: AchE: Acetylcholine esterase; WGS: Whole Genome Sequencing; AMR: Antimicrobial resistance; DRMs: Drug resistance mutations; MTB: M. tuberculosis; SA: S. aureus.
bAbbreviation: BWT: Burrows–Wheeler Transform, KMA: K-mer alignment, uses k-mer seeding to speed up mapping and the Needleman–Wunsch algorithm to accurately align extensions from k-mer seeds. BWA: Burrows-Wheeler Alignment, a short read alignment with BWT. MSA: multiple sequence alignment. TGH: A new statistical score, viz tree-generalized hypergeometric score. Kalign: An MSA program that uses a SIMD (single instruction, multiple data) accelerated version of the bit-parallel Gene Myers algorithm. MEM-Align: A fast semi-global alignment algorithm for short DNA sequences that allows for affine-gap scoring and exploit sequence similarity. BLAST: The Basic Local Alignment Search Tool.
cPerformance: The sample information of the performance corresponding to these severs is provided in detail. FastD: They detected 469 (89.7%) variants among the inserted variants, calling performance using AUC in ROC curve. ROC with an AUC of 0.870 indicated a reliable calling performance. They compared the detected allele frequencies of detected variants with their set allele frequencies and found that the allele frequencies calculated by FastD-TR were highly correlated with their ‘true’ allele frequencies (R2 = 0.834; ρ < 10−16). LRE-Finder: Fastq files from 21 LRE isolates were submitted to LRE-Finder. As negative controls, fastq files from 1473 non-LRE isolates were submitted to LRE-Finder. The MICs of linezolid were determined for the 21 LRE isolates. As LRE-negative controls, 26 VRE isolates were additionally selected for linezolid MIC determination. It was validated and showed 100% concordance with phenotypic susceptibility testing. PointFinder: A total of 685 different phenotypic tests associated with chromosomal resistance to quinolones, polymyxin, rifampicin, macrolides and tetracyclines resulted in 98.4% concordance. GWAMAR: Precision-recall curves for comparison of different association scores implemented in GWAMAR. One presents results for the mtu173 dataset (39 positives; 1450 negatives), AUC = 0.28; the other for the mtu_broad dataset (75 positives; 870 negatives), AUC = 0.43. Mykrobe predictor: With SE/SP of 99.1%/99.6% across 12 antibiotics (using an independent validation set, n = 470). For MTB, the method predicts resistance with SE/SP of 82.6%/98.5% (independent validation set, n = 1609). PhyResSE: PhyResSE was tested with 92 strains from a well-characterized strain collection from Sierra Leone that comprised 44 phenotypically susceptible strains and 48 strains. 100% concordance for resistance SNPs in katG, inhA, ahpC, rrs, rpsL, embA and embC; 98.91% concordance for those in gidB and pncA; and 97.83% concordance for those in rpoB and embB. KvarQ: KvarQ successfully detect all main DRMs and phylogenetic markers in 880 bacterial whole genome sequences. The variant calls of a subset of these genomes were validated with a standard bioinformatics pipeline and revealed >99% congruency. GenTB: using a ground truth dataset of 20,408 isolates with laboratory-based drug susceptibility data. The mean sensitivities for GenTB RF and GenTB-WDNN across the nine shared drugs were 77.6% and 75.4%, respectively. The specificity: GenTB-WDNN 96.2%, and GenTB-RF 96.1%. SAM-TB: The accuracy of SAM-TB in predicting drug-resistance was assessed using 3177 sequenced clinical isolates with results of phenotypic drug-susceptibility tests (pDST). Compared to pDST, the sensitivity of SAM-TB for detecting multidrug-resistant tuberculosis was 93.9% with specificity of 96.2%. Abbreviation: AUC: Area Under Curve. AC: Accuracy. SE: Sensitivity. SP: Specificity.
dSAM file: the file of SAM format; NGS: next generation sequencing.
This PDF is available to Subscribers Only
View Article Abstract & Purchase OptionsFor full access to this pdf, sign in to an existing account, or purchase an annual subscription.
