JuDFTteam / best-of-atomistic-machine-learning

🏆 A ranked list of awesome atomistic machine learning projects ⚛️🧬💎.
Creative Commons Attribution Share Alike 4.0 International
370 stars 35 forks source link

Sharpen the list's thematic focus #125

Closed Irratzo closed 1 month ago

Irratzo commented 12 months ago

Describe the issue.

The original and current thematic focus of this list is the application of machine learning methods to the domain of atomistic simulation. These include molecular modeling, most notably molecular dynamics, and electronic structure theory, most notably density functional theory. As the list is growing, it begins to touch on domains that have a topical overlap with this focus. The list should not include all areas of overlapping topics for three reasons.

  1. A lack of clear focus makes the list less relevant.
  2. The list maintainers cannot be experts in all connected domains.
  3. The currently used template has a fuzzy size limit of ca. 500 projects.

In that light, compile a list of domains with topical overlap to the current focus.

Related topics outside the list's focus.

Try to not include projects into the list that fall more into any of these domains than into the list's thematic focus.

The thematic focus of the list may shift. The list maintainers may change. Consequently, the decisions made here can change, too. These changes should be documented.

Irratzo commented 10 months ago

Actions for exclusion of biomolecules and computational life sciences.

Irratzo commented 1 month ago

In commit a960c2b, removed category biomolecules and associated projects UniFold, AlphaFold. Left label biomolecules in, though. There are still projects that either do not have a biomolecules focus, or they do, but may be used in other ways in atomistic ML from a chemistry / materials science perspective (e.g., the ZINC datasets).

The current disclaimer line about this thematic focus on materials and chemistry in config/header.md is okay, and does not have to be updated.

Create a new release from this update.