Dataset: "Joint structural annotation of small molecules using liquid chromatography retention order and tandem mass spectrometry data"

<p>Dataset used in the experiments of the publication: &quot;Joint structural annotation of small molecules using liquid chromatography retention order and tandem mass spectrometry data&quot; by Bach et al.</p> <p><strong>File description:</strong></p> <ul> <li> <p>cfmid4.tar: MS&sup2; spectra simulated using <a href="https://bitbucket.org/wishartlab/cfm-id-code/src/CFM-ID_4.0.7/">CFM-ID (v4.0.7)</a> for all molecular candidate structures</p> </li> <li> <p>db_layout.png: Visualization of the SQLite database (DB) layout</p> </li> <li> <p>massbank.sqlite.gz: DB containing all needed data to (re-)run the experiments shown in the paper. Please read &quot;DB_README.md&quot; for further details. The database file can be unpacked using gzip.</p> </li> <li> <p>metfrag.tar: MetFrag input files and MS&sup2; scores for all candidate sets computed using the <a href="https://ipb-halle.github.io/MetFrag/projects/metfragcl/">MetFrag software</a>.</p> </li> <li> <p>sirius_scores.tar: MS&sup2; scores for all candidates and measured spectra using the <a href="https://bio.informatik.uni-jena.de/software/sirius/">SIRIUS software</a>.</p> </li> <li> <p>sirius_inputs.tar: Input (ms-files) for the SIRIUS software.</p> </li> <li> <p>DB_README.md: Description of each table in the &quot;massbank.sqlite&quot; SQLite DB.</p> </li> <li> <p>db_processing_scripts.tar: Scripts to re-produce the &quot;massbank.sqlite&quot; and a README.md providing further information on the process.</p> </li> <li> <p>massbank__2020.11__v0.6.1.sqlite: Base SQLite DB from which the &quot;massbank.sqlite&quot; was build up. It was created using the &quot;<a href="https://github.com/bachi55/massbank2db">massbank2db</a>&quot; (v0.6.1) Python package using the <a href="https://github.com/bachi55/MassBank-data/tree/2020.11-branch">MassBank release 2020.11</a>.</p> </li> <li> <p>substructure_fingerprints.tar: Pre-computed substructure counting fingerprints for all candidates related to our experiments.</p> </li> </ul> <p><strong>Instructions:</strong></p> <p>The &quot;massbank.sqlite&quot; can be directly used with the Structure Support Vector Machine Model (SSVM) described in the manuscript and implemented in the &quot;<a href="https://github.com/aalto-ics-kepaco/msms_rt_ssvm">ssvm</a>&quot; Python package.</p> <p>If desired, the database can be re-produced using the scripts provided in &quot;db_processing_scripts.tar&quot;:</p> <ol> <li>Create a directory for all data</li> <li>Download and extract the ... <ol> <li>Processing scripts</li> <li>MS&sup2; scorer outputs (e.g. metfrag.tar)</li> <li>Pre-computed substructure fingerprints</li> </ol> </li> <li>Follow the instructions given in the &quot;README.md&quot; of the &quot;db_processing_scripts.tar&quot;</li> </ol>