tharos-ux / pancat

Pangenome graphs visualisation, distance computing, reconstruction of sequences and other utility functions

Welcome to Siegfried Dubois's Profile 👋

I'm programming in many languages, though most of my code here is written in Python.

• 👋 Hi, I’m Siegfried (He/Him)
• ❤️ I love algorithm creation, optimization and obfuscation.
• 🌱 I’m currently mainly a Python developper.
• 💼 I'm a PhD student at INRIA, in Brittany.
• 📖 I used to have courses about teaching and vulgarization.
• 🐲 I'm a roleplayer, some of my repos are linked to this hobby!
• 🎥 I do enjoy filmmaking, and have many ongoing projects.
• 🐱 I may like cats, also. Maybe a lil' too much. Meaow.

pancat's Issues

Add possibility to issue basic stats without plotting

As of now, to get metrics on the graph, one needs to compute and open the full graph. That's not the right way to do it.

Better display of node sizes in `grapher`

It has been determined the best way to inform the user of which size is the current segment is to use colors, but by classes (example: blue = 1-10bp, green = 10-50bp...)

Plooting a graph with new library leads to an error

AttributeError: 'Graph' object has no attribute 'colors' happens because new library was build without graphical concerns in mind. Need to add get_palette directly in pancat.

Loops are not taking care of by `offset`

As of now, a single offset per node per path is stored. However, we can cross a same node multiple times.

Get rid of "-g" parameter

Sniff the GFA format version in the header of the GFA file

Path selection with `isolate` is messy in case of loops

If a loop occurs, more than required is selected for the path, resulting in a messy GFA file.

pangraphs is not taking care of inversions

The commands reconstruct and edit are not aware of inversions. It will yield incorrect sequences, or refuse to do comparison because this is not correctly implemented.

Update tool to be compatible with new distribution of gfagraphs lib

require to go through each tool, change refs to data structure + imports

Loops within graphs extracted with `isolate` or `neighborhood` causes `grapher` to fall in infinite loop

When you work on a gfa graph that do has cycles, given the cycle is "long enough", the extraction of correct P or W lines fails, giving aberrant structures. Some nodes are referenced in P/W line, without being embed as S-lines, making the graph visualisation tool endlessly searching for those nodes.

Deleted dependency detected

I'm a Cyber Security researcher and developer of PackjGuard [1] to address open-source software supply chain attacks.

Issue

During my research, I found that this repo is vulnerable to attack due to deleted dependency from the public PyPI registry.

Details

Specifically, file requirements.txt lists gfatypes as one of the dependencies. However, it has been deleted from public PyPI. As such, an external bad actor can claim that name and register a malicious package, which will be then installed with pip install command, resulting in arbitrary remote code execution.

Impact

Not only your apps/services using https://github.com/Tharos-ux/pancat repo code are vulnerable to this attack, but the users of your open-source Github repo could also fall victim.

You could read more about such attacks here: https://medium.com/@alex.birsan/dependency-confusion-4a5d60fec610

Remediation

Please manually register a placeholder gfatypes package on PyPI immediately or remove gfatypes dependency from requirements.txt to fix this vulnerability.

To automatically fix such issues in future, please install PackjGuard Github app [1].

Thanks!

1. PackjGuard is a Github app that monitors your repos 24x7, detects vulnerable/malicious/risky open-source dependencies, and creates pull requests for auto remediation: https://github.com/marketplace/packjguard

Complete pangenome graph assertion

It has been suggested to add an option to reconstruct where you specifiy fasta input sequences, and the program tells if the graph is a complete pangenome graph or not.

Using `grapher` or `stats` on a graph without paths raises error

Error TypeError: 'builtin_function_or_method' object is not iterable is happening when one try to compute cyclicity of graph:

│ ❱ 84 │ stats["Is graph acyclic"] = all([len(set([x for x, _ in path_datas["path"]])) == len │
│ 85 │ │ path_datas["path"]) for path_datas in graph.paths.values]) │

Needs some default value and error handling if no paths is provided.

Fix `grapher` wrong path args

As of now, output is given as the form of path + job name. Would be better if user only gives the path to the output file.

BubbleGun issue with Cactus graph disables usage of command `neighborhood` on such graphs

Need to refactor the code to NOT use BubbleGun.