You probably have good intentions in asking this question but writing software for genomics is not a "one time and done" type thing. Biological data types keep changing/improving and the software needs to keep up/change. You will also need some understanding of basic biology and types of data.

If you are truly interested in long term contributions then there are opportunities to collaborate ( e.g. https://bioinformatics.ccr.cancer.gov/NCIBioinformaticsCommunity/collaboration-opportunities/ ). Rewriting existing software using more performant/secure languages (e.g. rust rewrites of existing software) is also an option (depending on your expertise and time availability).

I welcome these kinds of inputs into the community. There are, as you said, a lot of abandoned and obsolete projects in bioinformatics. That is why I am in favour of people doing rewrites in modern languages like Rust.

Here is a rewrite of the popular aligner minimap2 in rust - https://github.com/jwanglab/rammap. Turns out a significant bug was discovered that has existed for some time.

Many projects are in python2 or old perl and should be rewritten really. Tests are generally lacking across the board.

Examples of useful but obsolete projects (may be tricky licenses)

• https://github.com/Illumina/manta
• https://github.com/Illumina/strelka

Pangenomes are an area distinctly lacking in modern and fast tools, especially for viz and SNP calling - but it's hard.

• https://github.com/colindaven/awesome-pangenomes

You can check awesome lists and see if any devs are receptive to help.

• https://github.com/danielecook/Awesome-Bioinformatics
• https://github.com/seandavi/awesome-single-cell