Fabian Klötzl

In part one we used a switch statement to find the complement of a given base. Note how the nucleotides in the code are arranged in what looks like a table. However, this table is laid out in instructions space. Let's convert it to data space instead. char *revcomp_table(

When Watson and Crick described the two-stranded structure of DNA they stated that "if the sequence of bases on one chain is given, then the sequence on the other chain is automatically determined." This observation is crucial to much modern biology such as sequencing by synthesis. However, for a computer

Recently more and more CVEs are being filed against bioinformatic tools. Validity of the bug reports aside, it just shows that the security aspects of scientific software becomes more interesting. One of the most important factors is parsing and validating input data. Correctly handling invalid data is crucial to ensure

Mash could well be the new BLAST. It is a great technique to compute similarity scores for sequences. However, its implementation is not optimal. In this blog post I describe some tricks to increase the performance substantially. Addendum: Some people find this blog post and title rude. I hereby apologize

Parsing a FASTA file or any file format can be hard. But mostly writing a parser can be cumbersome. Thus, people don't do it properly or just copy someone elses work. For reading FASTA files a lot of bioinformatics tools use kseq.h. However, tool writers fail to get the

Genomes are very long strings. This comes to no surprise to anybody, except maybe a few C programmers. The following bug appears only in C programs or bad C++ programs due to the quirks of C. Strings in C are null-terminated and their length is implicit. This has two consequences:

“A QA Engineer walks into a bar. Orders a beer. Orders 0 beers. Orders 999999999 beers. Orders a lizard. Orders -1 beers. Orders a sfdeljknesv.” FASTA is the simplest file format ever invented by humans. However, the lack of a formal specification leads to high variation in the various implementations

RSPR is a tool to compute a topological distance between two phylogenies. It is an alternative to the Robinson-Foulds distance. The canonical implementation rspr has supports many different algorithms and options. Unfortunately, the authors decided not to use a standard way for parsing commandline arguments, such as getopt and instead

INDELible is one of the most sophisticated simulators for sequence evolution. It supports plenty of different sequence evolution models, various indel distributions, basically everything you could wish for. As such the code is pretty massive. It is also quite messy. The worst part: It contains six .cpp files which just

MUMmer is one of the most used sequence alignment tools, right after BLAST. Most of its code base is quite old but still being actively developed (see MUMmer4). The following issue has been sitting in there for over five years. struct Score { long int value; char used; }; struct Node { Score

This post will kick off a series of blog posts about the state of software engineering in bioinformatics. In each post I will take a stab at a popular tool in bioinformatics, dive in its code and show some of its worst parts. The issues will range from bad style,