SCHWED-6458: added ma-idk-u2osmap project.

projects/2025-01-ma-idk-u2osmap/README.md

+# Modelling of histone complexes (structural prediction screen)
+
+[Link to project in ModelArchive](https://www.modelarchive.org/doi/10.5452/ma-idk-u2osmap) (incl. background on project itself)
+
+Input files for conversion:
+- input_data_per_model.csv with metadata (incl. UniProtKB AC)
+- config.json file for ColabFold setup
+- .json & .pdb files for each model to convert. All .json files are expected to be listed in the CSV file. Expected file names are `[MDL_ID]_scores_rank_001_[AF_MDL_ID].json` and `[MDL_ID]_unrelaxed_rank_001_[AF_MDL_ID].pdb` with `MDL_ID` being two gene names as `[GN1]_[GN2]` and `AF_MDL_ID` being used to identify the AF2 model used (e.g. "multimer_v3_model_1_seed_000").
+
+Modelling setup:
+- Classic ColabFold setup with PDB coordinates and scores JSON and link to UniProt
+
+Special features here compared to the script used in `2024-08-ma-dm-hisrep`:
+- Improved parsing of UniProt versions by only downloading versions with different sequences and trying to find match with minimal mismatches and maximal coverage.
+- Added LocalColabFold as software.
+- Updates to support ModelCIF 1.4.7: QAMetricsFile instead of LocalPairwiseQAScoresFile.
+
+Content:
+- translate2modelcif.py : script to do conversion (run with same setup as Docker container here but with OST 2.9.2 and very latest main branch of python-modelcif and python-ihm from 4.3.2025)
+- modelarchive_minimal.zip: example input to convert a single complex from this set
+- modelcif_test.zip: output from running conversion of example with the command below : 
+```
+ost translate2modelcif.py ./modelarchive_minimal ./modelcif_test
+```
\ No newline at end of file