Biotite: a unifying open source computational biology framework in Python

By Patrick Kunzmann, K. Hamacher. 2018

128 citations recorded by the source corpus.

As molecular biology is creating an increasing amount of sequence and structure data, the multitude of software to analyze this data is also rising. Most of the programs are made for a specific task, hence the user often needs to combine multiple programs in order to reach a goal. This can make the data processing unhandy, inflexible and even inefficient due to an overhead of read/write operations. Therefore, it is crucial to have a comprehensive, accessible and efficient computational biology framework in a scripting language to overcome these limitations. We have developed the Python package Biotite: a general computational biology framework, that represents sequence and structure data based on NumPyndarrays. Furthermore the package contains seamless interfaces to biological databases and external software. The source code is freely accessible at https://github.com/biotite-dev/biotite. Biotite is unifying in two ways: At first it bundles popular tasks in sequence analysis and structural bioinformatics in a consistently structured package. Secondly it adresses two groups of users: novice programmers get an easy access to Biotite due to its simplicity and the comprehensive documentation. On the other hand, advanced users can profit from its high performance and extensibility. They can implement their algorithms upon Biotite, so they can skip writing code for general functionality (like file parsers) and can focus on what their software makes unique.

Subfields: Materials Chemistry, Molecular Biology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Materials Chemistry

1. Alexander Rives
2. Tom Sercu

In Molecular Biology

1. Brian L. Hie
2. Alexander Rives
3. Robert Verkuil
4. Adam Lerer
5. Chloe Hsu
6. Jason Liu
7. Tom Sercu
8. Zeming Lin

Citations in the corpus

• Evolutionary-scale prediction of atomic level protein structure with a language model (2022, neutral): “[63] Patrick Kunzmann and Kay Hamacher.”
• Learning inverse folding from millions of predicted structures (2022, methodology): “Solvent accessible surface area calculated with the Shrake-Rupley (“rolling probe”) algorithm from the biotite package (Kunzmann & Hamacher, 2018) and summed over all atoms in each amino acid.”
• Bilingual language model for protein sequence and structure (2024, unknown): “We used global Needleman-Wunsch alignment [47] as implemented in biotite [86] together with the 3Di substitution matrix from Foldseek [1] to compute sequence similarities.”