BioJava:Performance
BioJava performance examples
All tests can be run using Java Web Start
The full source code for all examples is available from the SVN repository
Read all chromosomes from Drosophila
Read all chromosomes from Drosophila and print out their sizes:
Run Example (download includes the 47MB file containing the genome sequence).
Results:
| System | Speed | Memory |
|---|---|---|
| Intel(R) Quad-Core Xeon @ 3GHz (Mac OS X 10.5.4, Java 6) | 9 sec. | 91 MB |
| Intel(R) Pentium(R) Dual CPU E2160 @ 1.80GHz (Linux, Java 6) | 16 sec. | 95 MB |
| Intel (R) Pentium (R) Dual CPU T2330 @ 1.60 GHz (Windows Vista, Java 6) | XX sec. | XX MB |
| Intel (R) Core 2 Duo @ 2.0GHz (Mac OS X 10.5.4, Java 6) | 16 sec | 81 MB |
| 1.33 Ghz PowerPC G4 (Mac OS X 10.4.9, Java 5) | 87 sec. | 81 MB |
The same example using the new BioJavaX code base (parses headers more thoroughly):
Run Example (download includes the 47MB file containing the genome sequence).
Results:
| System | Speed | Memory |
|---|---|---|
| Intel(R) Quad-Core Xeon @ 3GHz (Mac OS X 10.5.4, Java 6) | 7 sec. | 159 MB |
| Intel(R) Pentium(R) Dual CPU E2160 @ 1.80GHz (Linux, Java 6) | 16 sec. | 116 MB |
| Intel (R) Pentium (R) Dual CPU T2330 @ 1.60 GHz (Windows Vista, Java 6) | XX sec. | XX MB |
| Intel (R) Core 2 Duo @ 2.0GHz (Mac OS X 10.5.4, Java 6) | 14 sec | 199 MB |
| 1.33 Ghz PowerPC G4 (Mac OS X 10.4.9, Java 5) | 79 sec. | 108 MB |
Reverse complement of DNA sequence
Read DNA sequence and write their reverse complement. This is based on the benchmark provided at:http://shootout.alioth.debian.org/gp4/benchmark.php?test=revcomp&lang=all
read line-by-line a redirected FASTA format file.
for each sequence: write the id, description, and the reverse-complement sequence in FASTA format
Results:
| System | Speed | Memory |
|---|---|---|
| Intel(R) Quad-Core Xeon @ 3GHz (Mac OS X 10.5.4, Java 6) | 765 milli sec. | |
| Intel(R) Pentium(R) Dual CPU E2160 @ 1.80GHz (Linux, Java 6) | 1.1 sec | |
| Intel (R) Pentium (R) Dual CPU T2330 @ 1.60 GHz (Windows Vista, Java 6) | 1.5 sec. | |
| Intel (R) Core 2 Duo @ 2.0GHz (Mac OS X 10.5.4, Java 6) | 1.52 sec. | |
| 1.33 Ghz PowerPC G4 (Mac OS X 10.4.9, Java 5) | 4.4 sec |
Calculate structure alignment of Myoglobin and Haemoglobin
Calculate a protein structure alignment for Myoglobin (PDB code: 2jho) and Haemoglobin (PDB code: 2hhb). The matches to the 4 chains in Haemoglobin are identified as different alternate solutions.
Run Example (5MB download includes Jmol for visualization)
Results:
| System | Speed | Memory |
|---|---|---|
| Intel(R) Pentium(R) Dual CPU E2160 @ 1.80GHz (Linux, Java 6) | 4 sec. | < 100 MB |
| Intel (R) Pentium (R) Dual CPU T2330 @ 1.60 GHz (Windows Vista, Java 6) | 5 sec. | < 100 MB |
| Intel (R) Core 2 Duo @ 2.0GHz (Mac OS X 10.5.4, Java 6) | 8 sec | < 100 MB |
| 1.33 Ghz PowerPC G4 (Mac OS X 10.4.9, Java 5) | 26 sec. | < 100 MB |
Calculate a Sequence Alignment using Swith Waterman
Calculate a sequence alignment of two sequences of approx. 3000 nucleotides length (Corynebacterium renale plasmid pCR2, Pantoea agglomerans plasmid pPA3.0).
Results:
| System | Speed | Memory |
|---|---|---|
| Intel(R) Pentium(R) Dual CPU E2160 @ 1.80GHz (Linux, Java 6) | 5 sec | 129 MB |
| Intel (R) Pentium (R) Dual CPU T2330 @ 1.60 GHz (Windows Vista, Java 6) | 6 sec | 130 MB |
| Intel (R) Core 2 Duo @ 2.0GHz (Mac OS X 10.5.4, Java 6) | 4 sec | 120 MB |
| 1.33 Ghz PowerPC G4 (Mac OS X 10.4.9, Java 5) | 20 sec | 153 MB |