ACS Research Data Guidelines

TYPE OF DATA

The authors reporting analytical methods and data are encouraged to use best practices established in their respective fields of study and to adhere to official recommendations by the ACS, IUPAC or similar research body where possible. When such guidelines are used, please indicate this and cite the appropriate guidelines.

The data should be sufficiently transparent and rigorous to allow for the reproducibility of the experiments by others trained in the field and to evaluate the merits of the approach based on the available data given in the work.

Importantly, the analytical method should be critically evaluated in the intended complex sample. It should be cross-validated with an established reference technique when practically possible.

Appropriate analytical figures of merit measured in the complex sample of interest, where the sample characteristics are provided with sufficient detail to allow others trained in the field to reproduce the work. This includes data on reproducibility, accuracy, selectivity, sensitivity, detection limit and stability/lifetime so that others may be able to assess the performance of the method for the intended application.

SHARING DATA

ACS strongly endorses The FAIR Data Principles (Findable, Accessible, Interoperable, and Reusable). ACS supports the Center for Open Science’s Transparency and Openness Promotion (TOP) Guidelines, the Joint Declaration of Data Citation Principles, and the STM Brussels Declaration. ACS is a member of the Research Data Alliance.

For more information on these policies, please visit https://publish.acs.org/publish/data_policy

CHOOSING AN APPROPRIATE DATA REPOSITORY

Where possible, authors should deposit data in discipline-specific, community-recognized repositories or in generalist repositories when no community resource is available. ACS strongly encourages authors to select a repository that issues a persistent unique identifier, such as a DOI or an Accession Number. This will help facilitate the discoverability and citation of deposited data.

To find an appropriate repository, authors may refer to re3data.org and FAIRsharing.org for information on available repositories, their certification status, and services offered.

1. BIOLOGICAL SPECIMENS
   1. Antibodies: Authors are required to report the name of the antibody, the host species in which the antibody was produced and whether it is monoclonal or polyclonal. For commercial antibodies, report the company and catalog or code number and the antibody identifier obtained from The Antibody Registry. For academic antibodies, report the source laboratory and relevant reference. Clearly state the application for each antibody used in the manuscript. Include batch numbers for experiments in which variability is found among different antibody batches. Clearly state the final antibody concentration or dilution. Whenever possible, report the antigen or antigen location.
   2. Cell Lines and Microorganisms: To avoid inadvertent use of cross-contaminated or misidentified cell lines/microorganisms, authors are urged to validate each cell line/microorganism used. Authors must report the source of all cell lines/microorganisms in their manuscript, the date of authentication (must be within a year of manuscript submission date) and a description of the authentication method. Authors should be able to provide the authentication test results upon request. If no testing was done, provide the date when cells/microorganisms were purchased from authenticated source. For mammalian cell lines, authors must state whether the cell line has recently been tested for mycoplasma contamination. Resources for using cell lines as biological models:
      1. A resource for cell line authentication, annotation and quality control
      2. Cell Lines as Biological Models: Practical Steps for More Reliable Research
   3. Human Subjects: A statement confirming that the research has been approved by relevant ethical committees and performed under The Code of Ethics of the World Medical Association (Declaration of Helsinki) must be provided. Details listed in the latest version of the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines and description of informed consent protocols must also be provided. Authors reporting clinical trials should follow the CONSORT Statement for recommendations regarding the reporting of clinical trial results.
   4. Animal subjects Research involving animals must be performed in accordance with institutional guidelines as defined by the Institutional Animal Care and Use Committee for U.S. institutions or an equivalent regulatory committee in other countries. A statement confirming that all animal experiments performed for the manuscript were conducted in compliance with these guidelines is required. In the experimental section, the source, age, sex, species, and strain of animals should be included. For each treatment group, the number of animals used and sex should be clearly stated. Appropriate statistical methods should be used to test the significance of differences in results, and claims thereof. It is encouraged that all figure and table captions include the number of animals and sex for each treatment group, the method of statistical analysis as well as the corresponding p-values where significant differences are found.
      1. Further information on research involving animal and human subjects can be found in the following resources:
         • ACS Ethical Guidelines
      2. For key reagents and tools, we recommend the use of Research Resource Identifiers:
         • https://www.force11.org/group/resource-identification-initiative
   5. Biological Assays: Assay interference can cause misleading results. Thus, appropriate controls experiments should be performed to exclude common artefacts caused by reactive molecules (covalent and redox activity), colloidal aggregation, decomposition, and interference with the spectroscopic method. Authors should consult the recent ACS Editorial on assay interference compounds. The routes of administration of test compounds and vehicles should also be indicated. Benchmarks should be included in the form of appropriate positive or negative control substances or reference materials. Especially for studies on nanomaterials, additional steps and controls are needed such as sterilization procedures, checking assays for optical or chemical interferences, reporting different measuring units related to dose (e.g. surface area, mass, particle number per surface area, volume, cell number), and others as described in:
      1. ARRIVE (Animal Research: Reporting of In Vivo Experiments) guidelines
      2. Minimum information reporting in bio-nano experimental literature
2. SEQUENCE DATA

   Authors should submit sequence data to a public repository prior to submission and include accession numbers in their paper where appropriate.

   1. High-throughput sequencing data GEO
   2. DNA and RNA sequences: GenBank or NCBI Sequence Read Archive
   3. Nucleic acid sequencing data: NCBI Trace Archive or NCBI Sequence Read Archive (SRA)
   4. Protein sequences: Uniprot
3. COMPOUND CHARACTERIZATION

   Manuscripts should at least provide exemplary characterization and purity data for key compounds, including ¹H NMR, ¹³C NMR, and HRMS and preferably full characterization of all compounds described. Infrared absorbance (IR), specific optical rotation, and melting points should be included when appropriate. The purity and method of purity determination should be reported for all key compounds. For compounds prepared in a library format, a general experimental procedure should be provided, including full experimental details, with yields, for a representative selection of library members. The synthesis protocols and selected characterized compounds must reflect the reliability and scope of the reaction sequence. The purity of all reported library compounds should be explicitly stated. The submission of manuscripts purely based on mixture synthesis and/or mixture analysis is discouraged. Peptides should be characterized by HRMS and HPLC. For nanomaterials, the stability and reactivity of the nanomaterials as well as influence of external parameters like the composition of cell culture media, buffers etc. on nanomaterial properties should be addressed, for example by measuring dissolution, formation of reactive oxygen species or agglomeration under experimental conditions, and physical characterization techniques for nanomaterials should be sufficient, including description of algorithms and methods used to analyze the data. Please review the following resource:

   1. Minimum information reporting in bio-nano experimental literature

4. PROTEOMICS
   1. Sequence Analysis
      1. The method and/or program (including version number) used to create the "peak list" from the raw data and the parameters used in the creation of this peak list, particularly any which might affect the quality of the subsequent database search. Examples include whether smoothing was applied, any signal-to-noise criteria, whether charge states were calculated or peaks de-isotoped, etc. In cases where additional customized processing of the collections of peak lists has been performed, e.g. clustering or filtering, the method and/or program (including version number) should be referenced.
      2. The name and version of the program(s) used for database searching and the values of search parameters. Examples include precursor-ion mass tolerance, fragment-ion mass tolerance, modifications allowed for, any missed cleavages, protein cleavage chemistry, (if any), etc.
      3. The name and version of the sequence database(s) used. If a database was compiled in-house, a complete description of the source of the sequences is required. The number of entries actually searched from each database should be included. Authors should justify the use of a very small database or database that excludes common contaminants, since this may generate misleading assignments.
      4. Methods used to interpret MS/MS data, thresholds and values specific to judging certainty of identification, whether any statistical analysis was applied to validate the results, and a description of how applied.
      5. For large-scale experiments, provide the results of any additional statistical analyses that indicate or establish a measure of identification certainty, or allow a determination of the false-positive rate, e.g., the results of randomized database searches or other computational approaches.
   2. Information for each protein sequence identified
      1. Include accession number and database source
      2. Report score(s) and any associated statistical information obtained for searches conducted
      3. Report sequence coverage, expressed as the number of amino acids spanned by the assigned peptides divided by the sequence length
      4. Report the total number of peptides assigned to the protein
5. IMAGES
   1. Western Blots
      1. Manuscripts must report the primary antibody species, the secondary antibody species, isotypes, and generated epitopes. Catalog and lot numbers for commercially obtained antibodies must be reported, as well as blotting membrane and blocking agents.
      2. Full scans or images of uncropped blots must be provided
   2. Other images
      1. Original, uncropped images must be provided, at a minimum in the Supplementary Information. Any image manipulation must be described and justified in the text.
6. Biological Macromolecules from Electron Microscopy Experiments: Density maps should be deposited at either the Protein Data Bank in Europe (UK) or RCSB (USA) EMDB deposition site. Once the map has been deposited, any fitted atomic coordinates should be deposited with the Protein Data Bank (PDB) by following the link provided from the EMDB deposition session. The EMDB and PDB IDs should be included in the manuscript. Both the map and the coordinate data will be made public when the associated article is published.

GUIDELINES FOR REPORTING ELECTROCHEMICAL DATA AND PROCEDURES

Electroanalytical techniques are used to evaluate both molecules and materials for applications from energy storage and conversion to biosensors to electrosynthesis to fundamental characterization of a new molecule or material. Below is a list of guidelines for reporting procedures.

1. GUIDELINES FOR REPORTING VOLTAMMETRY AND AMPEROMETRY MEASUREMENTS

Voltammetry measurements such as cyclic (or linear or square wave) voltammetry are the basis for evaluating key reaction components (e.g., catalysts, substrates, etc.) and establishing the criteria for studying electrodes and electrocatalysts. Evaluated parameters should be included in the main text or supporting information (SI), so the experiments can be readily reproduced elsewhere.

Additionally, to support open data, uploading raw data is optional. If choosing to do so, the data need to be in a format that can be readily opened with a common data processing software such as MS Excel. Raw data can be deposited to a data repository with a DOI included in the Data Availability Statement section of the manuscript. The actual voltammograms should be included in the main text or the SI.

Please specify the following items along with the voltammogram (e.g., in the figure caption)

• CV plotting convention (i.e., polarographic or IUPAC).
• The material of the working electrode (WE), counter electrode (CE), and reference electrode (RE).
• Is the WE compartment separated from CE with a membrane or frit?
• Is the applied voltage corrected for IR compensation?
• If the working electrode is modified, then the procedure for the synthesis and modification should be included, including the substrate information (i.e., glassy carbon, carbon paper, carbon cloth).
• The solvent, electrolyte (including its concentration), analyte (including its concentration), and purging gas.
• The temperature (if not room temperature).
• The starting point of the scan, the direction of the scan, and scan rate.

Any modified electrode should include structural characterization of the electrode (i.e., SEM, XPS, TEM, etc.).

• For catalyst structural characterization, provide all data on the purity of catalysts, particle size distribution, and surface ligand coverage.
• Evidence should be provided for single-atom catalysts that the catalysts contain purely single atoms and are free of other clusters and nanoparticles.
• Catalyst characterization before and after electrochemical testing.
• For any electrochemical testing using Pt counter-electrodes where Pt contamination is a concern (i.e., testing a non-precious metal catalyst for comparison to a Pt catalyst), provide evidence for no Pt cross-contamination on active electrodes.

Please specify additional items in the Supporting Information when reporting cyclic voltammetry and other electrochemical procedures.

• The manufacturer and model of the electrochemical workstation.
• The material of the working electrode, counter electrode, and reference electrode (including electrolyte solution). The surface area and geometry of the working electrode. Include electrochemically active surface area (ECSA) for porous or nanostructured electrodes and the method employed for ECSA measurement.
• The polishing material and method.
• The solvent deoxygenation method, if applicable.
• The temperature.
• Stirring is not usually done during voltammetry measurements, but it should be noted if the solution is stirred or flowed through the cell, or the electrode is being rotated as in the case of a rotating disk electrode setup. Rotation rates should be included for rotating disk experiments.
• For electrocatalyst characterization, it is critical to report cyclic voltammograms in the presence and absence of the substrate reactant and the presence and absence of the catalyst. (i.e., for carbon dioxide reduction electrocatalysts, you should do an experiment of the electrolyte solution without CO₂ or catalyst, electrolyte solution with CO₂ but no catalyst, electrolyte solution with catalyst but no CO₂, and electrolyte solution with CO₂ and catalysts.) This is a required minimum set of experiments to determine that electrocatalysis is occurring. When possible, commercially available catalysts should be used as a benchmark, for example, Pt/C for HER, IrO₂ for OER, Cu/C for CO₂RR, etc. Other considerations:
  • Is the molecular mechanism(s) identified and verified?
  • If DFT calculation is presented, is the DFT model actually reflecting the same/real experimental system?

Necessary electrocatalyst experiments are discussed further in the bulk electrolysis section.

2. GUIDELINES FOR REPORTING BULK ELECTROLYSIS PROCEDURES

Please specify the following items in SI when reporting bulk electrolysis procedures.

• The manufacturer and model of the power supply or electrochemical workstation.
• The manufacturer, material, size, and geometry of the working electrode and counter electrode. As noted above, if the working electrode is a modified electrode, then the electrode modification procedure must be included.
• If a three-electrode system is employed, include the vendor of the reference electrode and the electrolyte solutions used.
• The fabrication procedure of any homemade electrode (reference, counter, or working).
• If using a commercial electrochemical cell, specify the manufacturer and model/catalog number. The dimensional parameter should be reported for homemade electrochemical cells. For H-cell, specify the material and specifications (e.g., pore size) of the separator being used.
• The distance between the working electrode and the counter electrode and the distance between WE and RE. The method employed to compensate IR.
• If the anode and cathode compartments are separated by a separator, provide the manufacturer or synthesis procedure for the separator as well as its thickness.
• If the reaction outcome is sensitive to stirring, include the size of the stir bar and stirring rate.
• If the reaction outcome is under flow conditions, include the flow rate.
• The temperature of the experiment and purging gas.
• If the reaction is carried out under galvanostatic (constant current), potentiostatic (constant potential), or constant cell voltage conditions, report the applied potential, current, or cell voltage. If the reaction is carried out under galvanodynamic or potentiodynamic conditions (such as alternating current electrolysis), report detailed parameters of the electrolysis conditions so that one could readily replicate the conditions with standard equipment.
• The electrolysis time. For organic electrochemistry, Faraday of electrons per mole of substrate (if readily calculated, such as under constant current or constant potential conditions) and product yields should be provided. For electrocatalytic measurements, Faradaic efficiency should be reported together with the total current density. Electrocatalytic stability measurements should be performed at the current density at which the reaction rate is reasonably high. When possible, current density should be normalized by the electrochemical surface area of the catalysts
• Clear images of the electrolysis setup.

For electrosynthetic experiments, details for product purification and data for compound characterization should be reported following the data requirement protocols of the target journal.

1. POLYMER CHARACTERIZATION

2. REPORTING RECOMMENDATIONS

1. Rheometry, Dynamic Mechanical Analysis (DMA), Mechanical Testing
   • The author should identify the make/model of any equipment used.
   • If performing materials testing guided by ASTM standards, this should be noted and the salient details reiterated.
   • The test geometry should be identified. For example:
     • If performing uniaxial tension measurements, the authors should specify such and additionally describe the dimensions of specimens used in testing.
     • If performing shear rheometry measurements, the authors should specify the plate diameter (upper and lower) and cone angle in the case of parallel plate/cone-and-plate geometry, dimensions in the case of concentric cylinder geometry, etc.
   • The processing steps taken to produce specimens and prepare them for measurement should be accurately documented.
     • Indicate if specimens were produced via melt-pressing, solvent casting, etc., including an accounting of the thermal/pressure/atmosphere history, and any other relevant information. If used as received, indicate as such
     • Further pre-treatment of specimens is often needed when mounting specimens into the measurement equipment. These additional processing steps should also be documented per the above.
     • In the case of shear rheometry, any pre-shear steps imposed on the sample (including pipetting) should be indicated, and the shear rates and equilibration times included.
     • The sample environment conditions (temperature/atmosphere) during measurement and pre-treatment should be described.
   • Any measurement parameters relevant to the specific experiment type should be documented. Appropriate representative data should be provided. For example:
     • Uniaxial Compression/tension test
       • Indicate the test speed in mm/min, %/min, or another reasonable permutation.
       • If samples were “preloaded” prior to test, the preload force should be identified.
       • It should be indicated whether “engineering stress/strain” or “true stress/strain” is measured, and this should be reported for all measured specimens (in SI). Averages of analyzed quantities of interest (modulus, toughness, etc.) with standard deviations and sample size should be reported.
     • Dynamic Mechanical Analysis (tension/compression/shear)
       • All Dynamic mechanical analysis experiments should report the storage modulus, loss modulus, and tangent (delta), for all measurement points. Additional raw data should be made available upon request
       • Indicate the pre-load force and application protocol for all tests
       • Dynamic strain sweep tests (f = 1 Hz) should be performed at representative temperatures/material states, results of which included in the SI, to show that any deriving experiments were performed in the Linear Viscoelastic Region (LVR).
       • Dynamic temperature ramp experiments should identify the deformation frequency, strain amplitude/strain application protocol, and temperature ramping rate.
       • Dynamic frequency sweep experiments should identify the range of deformation frequencies explored and the strain amplitude/strain application protocol. If multiple measurement temperatures were accessed, indicate the thermal process.
     • Flow Rheometry
       • Dynamic viscosity experiments should be carried out at discrete shear rates, the range of which indicated, with the steady-state equilibration and measurement time/protocol indicated. Representative time-dependent flow curves at relevant shear rates should be included in the SI to ensure that steady state is reached and accurately depicted.
       • The measured shear stress and calculated dynamic viscosity should be presented for every measured shear rate. Additional raw data should be made available upon request.
       • Indicate any pre-shear processing imposed on the sample prior to measurement.

2. Size Exclusion Chromatography

   Size exclusion chromatography (SEC) separates macromolecules on the basis of their size in solution or hydrodynamic volumes. This is a relative technique for determining the molar masses of polymer molecules. Thus, the system being used must be calibrated or must include a detector that is capable of absolute molar mass measurements such as an integrated, in-line light scattering detector. The basic minimum information that should be reported for SEC analysis is the following:

   (1) The type of pump, injector system, columns, and detectors being utilized. Include manufacturer information and relevant instrumental features (e.g., reference cell details, column packing details, laser wavelengths or scattering angles evaluated)
   (2) The mobile phase composition, flow rate, and temperature. Also report the temperature of the columns and the detectors.
   (3) The mass concentration and composition (i.e., if different from the mobile phase) of the analyte solution.
   (4) The calibrants used for analyzing the data (e.g., molar masses are reported relative to narrow molar mass distribution polystyrene standards)
   (5) Necessary sample characteristics relevant for data analysis (e.g., dn/dc values or Mark-Houwink constants)

   The above SEC metadata are important to include with raw data. When providing raw SEC data use comma separated variable (.csv) files that include elution volumes and corresponding detector responses.

   Typical SEC data are analyzed to give the mass average molar mass (M_m), the number average molar mass (M_n) and the dispersity (Đ = M_m / M_n) and those numbers should be reported for each polymer sample being analyzed. It should be clear how the molar mass of the analyte at each elution volume slice is being determined (e.g., relative to a standard or from a multiangle light scattering detector). It should also be clear how the baseline adjustments were performed (e.g., baseline flattening or subtractions) and the integration limits used for peak evaluations.

3. Differential Scanning Calorimetry (DSC)

   The basic minimum information that should be reported for DSC analysis
   • List the manufacturer and model of the instrument
   • List the sample mass and type of pan and lid used
   • If performing materials testing guided by ASTM standards, this should be noted and the salient details reiterated
   • Details of the complete thermal history or program should be provided, including any isothermal or temperature ramp steps. Duration and ramp rate, if applicable, of each step should be provided.
   • When determining the glass transition, melting, or crystallization temperature, the method of data analysis should be disclosed. For example, the author should disclose if the glass transition was taken as the inflection point, the half-height, or some other indication.
   • In reporting enthalpies, the method of integration and baselining should be disclosed
   • The number of scans performed and the specific scan from which the physical properties were analyzed should be disclosed (e.g., second heating scan, etc.).
   • The method of calibrating the DSC should be disclosed.
   • In the case of modulated DSC, the amplitude and period should be disclosed.
   • The type of gas in the DSC chamber should be reported.

4. Thermal Gravimetric Analysis (TGA)

   • List the manufacturer and model of the instrument
   • List the sample mass and type of crucible used
   • If performing materials testing guided by ASTM standards, this should be noted and the salient details reiterated.
   • Details of the complete thermal history or program should be provided, including any isothermal or temperature ramp steps. Duration and ramp rate, if applicable, of each step should be provided.
   • In reporting enthalpies, the method of integration and baselining should be disclosed.
   • The method of calibration should be disclosed.
   • The type of gas in the TGA chamber should be reported.

5. Light Scattering Analysis (DLS, SLS)

   The scattered intensity of polymer or colloidal solutions can be analyzed in dynamic (Dynamic Light Scattering, DLS) or static (Static Light Scattering, SLS) mode. The basic minimum information that should be reported for both DLS/SLS analysis is the following:
   • List the manufacturer and model of the instrument
   • Detail the sample cuvette used, solution medium, sample filtration, concentration of sample, temperature, number of measurements and duration
   • Model for data analysis should be provided
   • For DLS, the measured hydrodynamic radius (RH) or average diameter (z-average size) and polydispersity index (PDI) should be provided with the mode of analysis
   • For SLS, provide the method and model used to determine the radius of gyration (Rg), weight averaged molar mass (Mw) and second virial coefficient (A2), together with the range of angles used for data analysis

1. QUANTUM CHEMISTRY:

   Calculations should include coordinates or all key stationary points in a machine-readable format, such as .xyz, .mol2, or .pdb as appropriate. A table in the SI or a PDF are not acceptable sources for coordinates. Authors should make coordinate files available through ioChem-BD (www.iochem-bd.org), NOMAD (www.nomad-lab.eu), or another FAIR compliant repository. To find an appropriate repository, authors may refer to re3data.org and FAIRsharing.org for information on available repositories, their certification status, and services offered.

   Absolute energies of all key stationary points, as well as relevant vibrational frequencies should be provided as text within the body of the manuscript or the Supplementary Information.

   Inclusion of information on the normal modes, either directly or in the form of a Hessian matrix, is also encouraged where appropriate and relevant. This information can be supplied as text within the body of the manuscript or in the Supplementary Information.

   Quantum chemistry derived properties such as NMR shifts, key TD-DFT excitation energies, and oscillator strengths must also be supplied in a machine-readable format.

   In all cases, sufficient data must be supplied to reproduce the calculations. For calculations of molecular systems, this includes all keywords indicating method, basis set, solvation model, etc. The inclusion of the input file for each reported calculation, with indication of the package version used, is required. For periodic calculations of materials, information must include method, K-point list, and cutoffs. If widely distributed open-source or commercial software is used, full details identifying the specific version of software used must be provided. If in-house software is used, references must be provided to work that describes the software in detail and reports benchmark computations. In all cases a reader must be able to access the software and reproduce the calculations in the paper.

   Papers making use of quantum chemistry calculations must include a discussion of the method uncertainty, system size, DFT functional, and basis set.

2. SIMULATION

   All relevant simulation details should be described in the main text: the level of theory used, solvent model, integration method, software versions, and input parameters.

   Starting structures must be made available. Preference is for a link to a structure in the PDB or CCDC databases. Unpublished experimentally obtained structures should be deposited into appropriate repositories and referenced in the manuscript. If substantial modifications are made to an experimental structure prior to simulation, coordinates of the modified structures should be provided. Structure preparation, including solvation, and equilibration steps, must be described with sufficient detail to be reproduced.

   Simulation software must be described, including version and any plug-ins used. Simulation input files must be provided. In cases where a large number of simulations were performed, a template input file should be provided along with a description of the changes for each individual run, with enough detail for the entire set of simulations to be reproduced. Any adjustments to standard parameters must be described, and the modified parameters provided.

   Output trajectories must be provided with an appropriate level of detail. Full trajectories are required except in such cases for which the full trajectory is prohibitively large, in which cases, at a minimum, snapshots along the trajectory with solvent/counterions included if used. For free energy calculations, energy files should also be included. For simulations with large outputs, authors are encouraged to discuss the level of required trajectory detail with the Editor.

   Statistical uncertainties for all properties calculated from a trajectory must be described, and a description of how these were determined must be provided

3. MACHINE LEARNING
   1. Data sources

      Data sources must be listed and must be publicly available. Guidance on acceptable repositories is provided in the ACS Research Data Policy. If data is stored in an external database, an access date or a version number must be provided. Authors should discuss in the manuscript text any potential biases in the source data, as well as what approaches were used to mitigate any bias.

   2. Data Cleaning

      Data cleaning steps must be clearly and fully described, either in the text or as a code pipeline in publicly accessible code.

      The amount of source data removed must be explicitly listed and evaluated.

      Any instances of combining data from multiple sources must be clearly identified, with mitigation of potential issues discussed.

   3. Data Representations

      The methods for representing data as features or descriptors must be clearly articulated, with software implementations.

      Comparisons against standard feature sets must be provided.

   4. Model Choice

      The implementation of models used must be provided such that it can be trained and tested with new data.

      Baseline comparisons to simple/trivial models must be provided (e.g. 1-nearest neighbor, random forest, most frequent class).

      Baseline comparisons to current state-of-the-art must be provided.

   5. Model Training and Validation

      The model must clearly split data into different sets for training (model selection), validation (hyperparameter optimization), and testing (final evaluation).

      The method of splitting data must be clearly stated, and mimic the anticipated real-world application.

      The data splitting procedure must avoid leakage (e.g. the same composition present in training and test sets).

   6. Code and Reproducibility

      The code or workflow must be available in a public repository. GitHub is preferred, but other repositories compliant with FAIR principles are also acceptable. See the ACS Data Policy guidance on choosing appropriate repositories. Any scripts used to produce findings in the paper should be provided in full.