Coverage Bias and Sensitivity of Variant Calling for Four Whole-genome Sequencing Technologies
Bibliographic record
Abstract
The emergence of high-throughput, next-generation sequencing technologies has dramatically altered the way we assess genomes in population genetics and in cancer genomics. Currently, there are four commonly used whole-genome sequencing platforms on the market: Illumina's HiSeq2000, Life Technologies' SOLiD 4 and its completely redesigned 5500xl SOLiD, and Complete Genomics' technology. A number of earlier studies have compared a subset of those sequencing platforms or compared those platforms with Sanger sequencing, which is prohibitively expensive for whole genome studies. Here we present a detailed comparison of the performance of all currently available whole genome sequencing platforms, especially regarding their ability to call SNVs and to evenly cover the genome and specific genomic regions. Unlike earlier studies, we base our comparison on four different samples, allowing us to assess the between-sample variation of the platforms. We find a pronounced GC bias in GC-rich regions for Life Technologies' platforms, with Complete Genomics performing best here, while we see the least bias in GC-poor regions for HiSeq2000 and 5500xl. HiSeq2000 gives the most uniform coverage and displays the least sample-to-sample variation. In contrast, Complete Genomics exhibits by far the smallest fraction of bases not covered, while the SOLiD platforms reveal remarkable shortcomings, especially in covering CpG islands. When comparing the performance of the four platforms for calling SNPs, HiSeq2000 and Complete Genomics achieve the highest sensitivity, while the SOLiD platforms show the lowest false positive rate. Finally, we find that integrating sequencing data from different platforms offers the potential to combine the strengths of different technologies. In summary, our results detail the strengths and weaknesses of all four whole-genome sequencing platforms. It indicates application areas that call for a specific sequencing platform and disallow other platforms. This helps to identify the proper sequencing platform for whole genome studies with different application scopes.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".