Platform | Throughput | Length | Quality | Cost | Applications | Sources of error | Advantages | Disadvantages |
---|---|---|---|---|---|---|---|---|
Sanger | 6 Mb/day | 1,000 nt | 10-4-10-5 | ~$500/Mb | Small sample sizes, genomes, SNPs, long haplotypes, low complexity regions, etc. | Polymerase/amplification, low intensities/missing termination variants, contaminant sequences | Longest reads, gold standard for validations | High cost, low throughput |
454/Roche | 750 Mb/day | 400 nt | 10-3-10-4 | ~$20/Mb | Complex genomes, SNPs, structural variation, indexed samples, small RNAs, mRNAs, etc. | Amplification, mixed beads, intensity thresholding, homopolymers, phasing, neighbor interference | Longer reads, easier to assemble | Medium throughput, expensive, indel errors more likely |
Illumina | 5,000 Mb/day | 100 nt | 10-2-10-3 | ~$0.50/Mb | Complex genomes, counting (SAGE, CNV Chip, small RNA), mRNAs, structural variation, bisulfite data, indexing SNPs, etc. | Amplification, mixed clusters/neighbor interference, phasing, base labeling | Lower cost, widely adopted platform, most well-developed bioinformatics efforts | Higher base substitution error rate, shorter reads |
SOLiD | 5,000 Mb/day | 75 nt | 10-2-10-3 | ~$0.50/Mb | Complex small genomes, counting (SAGE, ChiP, small RNA, CNV), SNPs, mRNAs, structural variation, indexing, etc. | Amplification, mixed beads, phasing, signal decline, neighbor interference | Lower cost, 2-base encoding chemistry, higher per-base accuracy | Shortest read lengths, still an emerging platform |