We use GMGC, HIV, CoVID-19 and MPoX datasets to measure our methods.
Citations:
- GMGC: Coelho, L.P., et al. Towards the biogeography of prokaryotic genes. Nature 601, 252–256 (2022).
- HIV: https://www.hiv.lanl.gov/
- CoVID-19: Shu YL, McCauley J. GISAID: Global initiative on sharing all influenza data - from vision to reality. Eurosurveillance. 2017; 22(13):2-4.
- MPoX: Ma Y, Chen M, Bao Y, et al. MPoxVR: A comprehensive genomic resource for monkeypox virus variant surveillance. The Innovation, 2022, 3(5).
The generated test data is saved in test_case.tar.xz. This data also saved at Zenodo. The following command provides the way to decompress the file:
mkdir test_case
tar xvf test_case.tar.xz -C ./test_caseFor every file in test_case folder, the name of file is the source of test cases (e.g., GMGC_20000__GMGC_10000 means the file is generated by test case GMGC_20000 and GMGC_10000 for pairwise sequence alignment, the first sequence is randomly selected in GMGC_20000 file, the second sequence is randomly selected in GMGC_10000 file), the *.ans file is CIGAR string generated by Benchmark (i.e., gap-affine-swg in WFA2).
Test programs has compiled in C++20, and the program is static linked (i.e. no extra dependencies, can be runned in any Linux platforms). The programs has stored in wfa2-test-prog folder.
All tests are tested in Linux. The results are generated by align_benchmark and multiswg_benchmark programs compiled by wfa2-test folder in FORAlign.
The organized results is saved in test-result/json folder. The following table shows the content for these datas:
| File Name | Content Info | Used in test |
|---|---|---|
| 9gpu_*core.json | The test result from 9gpu device use the determined number of cores | all of tests |
| 9gpu_benchmark.json | gap-affine-swg result generated by 128 cores |
benchmark for anaylzing datas |
| hpc_56core_cpucnt_anaysis.json | The test result which the CPU percent tested by top is more than 95% generated by 56 cores |
condition_variable and Speedup Ratio for swg |
| hpc_56core_simpletest.json | The test result which the CPU percent tested by top is more than 95% generated by 56 cores, not tested WFA series |
t-blocks |
| hpc_80core_cpucnt_anaysis.json | The test result which the CPU percent tested by top is more than 95% generated by 80 cores |
all of tests except Speedup Ratio for swg |
| hpc_80core_max_memory.json | The maximum memory in Benchmark methods |
Methods selection |
| singlepc_1_12core.json | The test result generated by 12 cores |
all of tests except Speedup Ratio for swg |
| singlepc_2_16core.json | The test result generated by 16 cores |
supplemental data |
| singlepc_2_24core.json | The test result generated by 24 cores |
all of tests except Speedup Ratio for swg |
| 9gpu_256core_Alignment_Argument.json | The test result generate for measuring alignment argument | alignment parameter settings in FORAlign section |
| bowtie2_res.json | The test result between FORAlign, WFA and Bowtie2 | Comparison between FORAlign, WFA and Bowtie2 section |
The raw test cases is in test-result/raw.tar.xz. The following command provides the way to decompress the file:
tar xvf test-result/raw.tar.xzFiles will in test-result/raw folder.