Table 4.
Neoantigen prediction softwares.
| Software (references) | Principle | Year |
|---|---|---|
| NeoPredPipe [140] | Connects commonly used bioinformatics software using custom python scripts giving neoantigen burden, immune stimulation potential, tumor heterogeneity and HLA haplotype of patients. | 2019 |
| Strelka2 [141] | Estimates error or deletion parameters of each sample improved tumor liquid analysis | 2018 |
| MuPeXI [142] | Identifies tumor-specific peptides through the extraction and induction of mutant peptides, it can predict immunogenicity and evaluate the potential of novel peptides | 2017 |
| CloudNeo pipeline [143] | The docker container executes the tasks. After giving as an input mutant VCF file and bam FILE representing HLA typing, the software predicts HLA affinity all mutant peptides. | 2017 |
| pVAC-Seq [144] | Integrates tumor mutation and expression data to identify personalized mutagens through personalized sequencing. | 2016 |
| NetMHCpan [145] | The sequences are compared using artificial intelligence neural network and predict affinity of molecular peptide-MHC-I type | 2016 |
| VariantEffect Predictor Tool [146] | It uses automated annotations to manual review time and prioritize variants | 2016 |
| Somaticseq [147] | It uses a randomized enhancement algorithm, which has more than 70 individual genome sequence features based on candidate sites to accurately detect somatic mutations | 2015 |
| OptiType [148] | It uses an HLA type algorithm with a linear programming that gives sequencing databases comprising RNA, exome and whole genome sequencings. | 2014 |
| ATHLATES [149] | It assembles allele recognition, pair interface applied to short sequences and HLA genotyping at allele level achieved via exon sequencing | 2013 |
| VarScan2 [150] | It detects somatic and copy number mutations within tumor-normal exome data using a heuristic statistical algorithm. | 2012 |
| HLAminer [151] | Through a shotgun sequencing Illumina database platform, predicts HLA type through an orientation of the assembly of the shotgun sequence data to then compare it with databases of allele sequences used as references. | 2012 |
| Strelka [152] | It uses a Bayesian model that matches normal-tumor sample sequencing data to analyze and predict with high accuracy and sensitivity somatic cellular variations | 2012 |
| SMMPMBEC [153] | Through a Beyesian matrix based on optimal neural network they can predict peptide molecules with MHC-I | 2009 |
| UCSC browser [154] | The fusion of various databases can give fast and accurate access to any gene sequence. | 2002 |