Progress in quickly finding orthologs as reciprocal best hits: comparing blast, last, diamond and MMseqs2
Bibliographic record
Abstract
BACKGROUND: Finding orthologs remains an important bottleneck in comparative genomics analyses. While the authors of software for the quick comparison of protein sequences evaluate the speed of their software and compare their results against the most usual software for the task, it is not common for them to evaluate their software for more particular uses, such as finding orthologs as reciprocal best hits (RBH). Here we compared RBH results obtained using software that runs faster than blastp. Namely, lastal, diamond, and MMseqs2. RESULTS: We found that lastal required the least time to produce results. However, it yielded fewer results than any other program when comparing the proteins encoded by evolutionarily distant genomes. The program producing the most similar number of RBH to blastp was diamond ran with the "ultra-sensitive" option. However, this option was diamond's slowest, with the "very-sensitive" option offering the best balance between speed and RBH results. The speeding up of the programs was much more evident when dealing with eukaryotic genomes, which code for more numerous proteins. For example, lastal took a median of approx. 1.5% of the blastp time to run with bacterial proteomes and 0.6% with eukaryotic ones, while diamond with the very-sensitive option took 7.4% and 5.2%, respectively. Though estimated error rates were very similar among the RBH obtained with all programs, RBH obtained with MMseqs2 had the lowest error rates among the programs tested. CONCLUSIONS: The fast algorithms for pairwise protein comparison produced results very similar to blast in a fraction of the time, with diamond offering the best compromise in speed, sensitivity and quality, as long as a sensitivity option, other than the default, was chosen.
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".