DCHO: Context-Dependent Interpretations

• DCHO is a multifaceted term with distinct meanings in astrochemistry, cultural heritage informatics, and neuroimaging, each relying on domain-specific definitions and applications.
• In astrochemistry, DCHO is commonly used as an informal label for deuterated species, where HDCO serves as the standard notation for singly deuterated formaldehyde and precise isotopologue specification is required.
• In cultural heritage and neuroimaging, DCHO refers respectively to a Digital Cultural Heritage Object produced via detailed processing workflows and a decomposition–composition framework for forecasting higher-order brain connectivity.

Searching arXiv for papers using the query “DCHO”. DCHO denotes distinct entities in different research literatures rather than a single stable concept. In astrochemistry, it appears as a query term for deuterated species but the formal notation in the cited literature is usually HDCO for singly deuterated formaldehyde, while deuterated acetaldehyde is resolved into explicit isotopologues such as CH$_2$DCHO, CH$_3$CDO, and CHD$_2$CHO. In cultural heritage informatics, DCHO means Digital Cultural Heritage Object. In neuroimaging, DCHO names a Decomposition–Composition framework for Higher-Order brain connectivity (Bergman et al., 2010, Asensio et al., 2023, Barzaghi et al., 2024, Li et al., 27 Aug 2025).

1. Domain-dependent meanings

The term is best understood through its disciplinary expansions and the object to which each expansion refers.

Domain	Meaning of DCHO	Representative source
Astrochemistry of formaldehyde	Query wording for singly deuterated formaldehyde; paper notation uses HDCO	(Bergman et al., 2010)
Astrochemistry of acetaldehyde	Loose shorthand for a deuterated acetaldehyde, chemically ambiguous without isotopologue specification	(Asensio et al., 2023)
Cultural heritage informatics	Digital Cultural Heritage Object	(Barzaghi et al., 2024)
Neuroimaging	Decomposition–Composition framework for Higher-Order brain connectivity	(Li et al., 27 Aug 2025)

Within astrochemistry, the key terminological fact is that the formal spectroscopic literature is more specific than the query term. The $\rho$ Oph A study explicitly states that it uses HDCO, not “DCHO,” for singly deuterated formaldehyde, and further notes that HDCO and DCHO refer to the same isotopologue in astrochemical terms (Bergman et al., 2010). By contrast, the acetaldehyde spectroscopy paper states that “DCHO” is chemically ambiguous because it may point either to methyl-side deuteration, most commonly CH$_2$DCHO, or to aldehydic deuteration, CH$_3$CDO, whereas that work concerns CHD$_2$CHO (Asensio et al., 2023).

Outside astrochemistry, DCHO is not a molecular label at all. In the Aldrovandi Digital Twin workflow it is an operational data-management category, and in the higher-order brain connectivity paper it is the name of a predictive model architecture (Barzaghi et al., 2024, Li et al., 27 Aug 2025).

2. DCHO as singly deuterated formaldehyde: the HDCO literature

When DCHO is used in the sense of deuterated formaldehyde, the relevant spectroscopic symbol is HDCO. The $\rho$ Ophiuchi A mapping study analyzed H$_2$CO, HDCO, and D$_2$CO over the central $_3$0 region of $_3$1 Oph A at a distance of about $_3$2 pc, using the 12 m APEX telescope in the 1.3 mm band. The mapped HDCO emission is dominated by the $_3$3 line at 246924.60 MHz, while the lower $_3$4 line at 227668.45 MHz was not detected because of its very low $_3$5-coefficient. The integrated intensity of the detected HDCO line is $_3$6 at the D-peak $_3$7, where the derived column density is $_3$8 and the deuteration ratio is $_3$9. At the same position, $_2$0, reported as the first case in which D$_2$1CO is more abundant than HDCO. The authors infer kinetic temperatures of roughly 20–30 K and densities of $_2$2 for the cores, and they argue that depleted gas-phase chemistry is inadequate, favoring grain-surface abstraction and exchange along $_2$3 (Bergman et al., 2010).

The same isotopologue is used as an evolutionary tracer in massive star-forming regions. In that survey, the singly deuterated species is again treated as HDCO, with APEX SEPIA Band 5 observations of several $_2$4 lines and ALMA follow-up on $_2$5 scales. HDCO is detected toward high-mass protostellar objects (HMPOs) and ultracompact H II regions (UC HII regions), but not toward the observed high-mass starless cores (HMSCs). The paper’s principal result is that the formaldehyde deuteration fraction decreases by about an order of magnitude from earlier HMPO-like phases to the UC HII stage, from $_2$6 to $_2$7. The authors use the Rodgers–Charnley parameter

$_2$8

and conclude that the observed values are not well explained by pure gas-phase chemistry and are more consistent with grain-surface chemistry, while also emphasizing that the HMSC phase remains poorly constrained because the upper limits are high (Zahorecz et al., 2021).

High-resolution ALMA work on NGC 1333 IRAS 4A further sharpens the diagnostic role of the HDCO–D$_2$9CO pair. There, the relevant measured quantity is $\rho$0 along the cavity walls of the outflows from IRAS 4A1. The ratio declines from roughly 20–60% at projected distances of about 2000 au to about 10% at 4000 au, and the authors interpret both HDCO and D$\rho$1CO as grain-surface products formed in the prestellar phase and later injected into the gas by shocks. The resulting gradient is proposed to trace the original prestellar density structure, with a relatively flat density profile inside $\rho$2 au and decreasing density beyond that radius (Chahine et al., 2024).

Across these studies, the central misconception to avoid is that “DCHO” is a distinct formal astrochemical symbol. In the formaldehyde literature represented here, the standard isotopologue notation is HDCO, and its scientific meaning emerges mainly through comparison with H$\rho$3CO and D$\rho$4CO rather than through the query label itself (Bergman et al., 2010, Zahorecz et al., 2021).

3. DCHO in deuterated acetaldehyde spectroscopy

In acetaldehyde studies, “DCHO” does not uniquely identify a molecule. The millimetre and sub-millimetre spectroscopy paper on CHD$\rho$5CHO states explicitly that “DCHO” is often used loosely for a deuterated acetaldehyde but is chemically ambiguous, because it may refer to CH$\rho$6DCHO or CH$\rho$7CDO, whereas the paper itself focuses on the doubly deuterated methyl isotopologue CHD$\rho$8CHO (Asensio et al., 2023).

That work combines laboratory spectroscopy with astrophysical detection. Rotational transitions were measured from 82.5 to 450 GHz, and a global fit reproduced 853 transition frequencies with a weighted root mean square standard deviation of 1.7, varying 40 spectroscopic constants. The molecule is treated as a non-rigid rotor with hindered internal rotation of an asymmetrical CHD$\rho$9 methyl group, using an effective Hamiltonian based on the high-barrier internal axis method. The fitted model distinguishes two equivalent low-energy Out conformers and one higher-energy In conformer, with the derived zero-point-energy difference $_2$0. The resulting catalogue is intended for astronomical identification and includes predicted frequencies, uncertainties, $_2$1 intensities at 300 K, lower-state energies, upper-state degeneracies, and full quantum assignments (Asensio et al., 2023).

The catalogue enabled the first interstellar detection of CHD$_2$2CHO toward IRAS 16293-2422, specifically the B component, using the ALMA Protostellar Interferometric Line Survey. The analysis assumed LTE with $_2$3, FWHM = 1 km s$_2$4, and a $_2$5 offset of 2.6 km s$_2$6. The derived column density is $_2$7 with 10–20% uncertainty, and the corresponding doubly-to-singly deuterated ratio is approximately $_2$8. The paper notes that this D$_2$9/D ratio is similar to those measured for other complex organic molecules toward IRAS 16293B, including methyl formate, dimethyl ether, and methanol (Asensio et al., 2023).

For the interpretation of “DCHO,” the significance of this paper is terminological as much as spectroscopic. It shows that any chemically precise use of the label in acetaldehyde chemistry must resolve the isotopologue explicitly, because the position and multiplicity of deuteration are part of the scientific content rather than a minor naming detail (Asensio et al., 2023).

4. DCHO as Digital Cultural Heritage Object

In cultural heritage informatics, DCHO has no chemical meaning. It stands for Digital Cultural Heritage Object and is a core operational term in the Aldrovandi Digital Twin workflow. The paper distinguishes the physical cultural heritage object (CHO) from the digital object produced through digitisation, the DCHO, and also distinguishes an optimized online-delivery derivative, DCHOo (Barzaghi et al., 2024).

The workflow is explicitly staged. Acquisition captures the physical CHO and produces RAW data. Automatic processing converts RAW into RAWp, a first processed raw model. The Modelling phase (step 3) is where a human operator resolves topological issues in RAWp and thereby obtains the DCHO. The Optimisation phase (step 4) simplifies the DCHO for specific use cases and yields DCHOo. Export then converts RAWp, DCHO, and DCHOo into concrete formats, with OBJ or FBX used for RAWp and DCHO, MTL associated with OBJ exports, textures in PNG or JPG, and glTF used for DCHOo. In the final Upload phase (step 7), DCHOo is published through the web-based framework ATON (Barzaghi et al., 2024).

The paper treats the DCHO as a full digital research object with its own metadata, provenance, identifiers, and preservation strategy. FAIR implementation is organized at three levels: objects, metadata about the object, and metadata records. The semantic and provenance stack includes RDF, OWL ontology, Linked Open Data, SPARQL endpoint, CIDOC CRM, CRMdig, CHAD-AP, and OCDM. Object-level provenance is described as OPI (Object Provenance Information), and provenance of the metadata itself as MRPI (Metadata Record Provenance Information). The repository plan is to deposit 3D models and metadata in Zenodo, obtain DOIs, and potentially migrate later to infrastructure emerging from H2IOSC (Barzaghi et al., 2024).

A particularly concrete contribution of the paper is its treatment of storage and lifecycle complexity. In the reported case study, occupied storage is distributed as 46% RAW, 43% RAWp, 7% DCHO, <1% DCHOo, and 4% documentation. The authors stress that the expensive part of 3D heritage preservation is therefore not the dissemination-ready optimized model but the preservation-worthy upstream material. They also emphasize that 3D heritage data are more difficult to manage than images because they involve multiple object states, file-format fragmentation, larger storage burden, loss of metadata during conversion, limited publication infrastructures, and unresolved questions about the interpretive and legal status of the resulting digital objects (Barzaghi et al., 2024).

Within this literature, DCHO is thus neither a generic 3D file nor a mere surrogate of the original object. It is the modeled digital heritage object produced after processing and human-led modeling, situated within a FAIR-by-design pipeline (Barzaghi et al., 2024).

5. DCHO as a Decomposition–Composition framework for higher-order brain connectivity

In neuroimaging, DCHO is the title of a computational framework rather than an object or isotopologue. The acronym refers to a Decomposition–Composition framework for Higher-Order brain connectivity, designed to infer and forecast the temporal evolution of higher-order brain connectivity (HOBC) from fMRI-derived data (Li et al., 27 Aug 2025).

The motivation of the framework is that conventional functional connectivity (FC) captures only pairwise interactions, whereas HOBC is intended to represent interactions among three or more brain regions. In the paper, the focus is on third-order interactions, encoded as a weighted simplicial structure and concretely stored as a tensor

$_3$0

where $_3$1 is the weighted higher-order co-fluctuation among regions $_3$2 at time $_3$3. The central modeling move is to avoid direct prediction of future HOBC tensors from raw inputs. Instead, the task is decomposed into HOBC inference and latent trajectory prediction, and then recomposed through decoding (Li et al., 27 Aug 2025).

The architecture has two major stages. In the inference stage, a dual-view encoder combines a local topological extractor with a global topological extractor based on spectral graph convolution with Chebyshev polynomials. The resulting latent representation is passed to a higher-order decoder that contains a latent combinatorial learner and a dual-stream Transformer with spatial and temporal branches. In the forecasting stage, a multilayer LSTM predicts future latent trajectories, and a latent-space prediction loss

$_3$4

is used instead of direct HOBC-space supervision. The paper also provides a theorem stating that future HOBC error is bounded by the sum of the inference error and a decoder-Lipschitz-scaled latent prediction error, thereby justifying the decomposition strategy (Li et al., 27 Aug 2025).

Implementation details are fully specified. The framework is implemented in PyTorch. The encoder has two parallel branches, each a two-layer Graph Network. The latent dimension is 32. The predictor is a 4-layer LSTM. The spatial and temporal branches of the higher-order decoder are two-layer TransformerEncoders with 4 attention heads, feedforward dimension equal to 2 $_3$5 latent size, and GELU activation. Optimization uses Adam with learning rate $_3$6, weight decay $_3$7, dropout 0.1, 200 epochs, and batch size 8. Each result is repeated 10 times and reported as mean $_3$8 standard deviation (Li et al., 27 Aug 2025).

The evaluation covers Human Connectome Project datasets—Emotion, Gambling, Language, Motor, Relational, Social, Working Memory, and Rest—as well as synthetic Lorenz and Hindmarsh–Rose systems. For 10-step HOBC forecasting, DCHO outperforms MLP, LSTM, and Transformer baselines on all reported datasets and metrics. For example, on Emotion the model reaches MAE 0.1744 and RMSE 0.2672, compared with 0.2887 and 0.4709 for the best listed baseline LSTM; on Rest it reports MAE 0.1704 and RMSE 0.2738, compared with 0.3865 and 0.5240 for LSTM. The framework also leads raw fMRI forecasting and improves task-state classification, with reported average gains of +6.24% Accuracy, +6.40% Precision, +5.58% Recall, +6.00% F1, and +6.77% AUROC over the compared representations. Ablation experiments show major degradation when removing the decomposition strategy, the latent combinatorial learner, or the latent-space prediction loss (Li et al., 27 Aug 2025).

Here the meaning of DCHO is therefore entirely algorithmic: it denotes a specific higher-order representation-learning and forecasting pipeline, not a data type and not a molecule (Li et al., 27 Aug 2025).

6. Disambiguation, notation, and cross-domain significance

The most important general point is that DCHO is a context-bound term. In astrochemistry, the same string may refer informally to deuterated formaldehyde or to a deuterated acetaldehyde, but in both cases chemically explicit notation is preferred in the primary literature. HDCO is the standard notation for singly deuterated formaldehyde in the cited formaldehyde papers, while acetaldehyde work distinguishes CH$_3$9DCHO, CH$_2$0CDO, and CHD$_2$1CHO because isotopologue-resolved chemistry and spectroscopy are central to the scientific interpretation (Bergman et al., 2010, Asensio et al., 2023, Zahorecz et al., 2021).

In cultural heritage, DCHO is instead part of a lifecycle ontology. It is the modeled digital object obtained after RAW acquisition, processing, and human-guided correction, and it remains distinct from both the original CHO and the optimized dissemination derivative DCHOo. In neuroimaging, the same four letters identify a model family organized around latent HOBC inference and prediction (Barzaghi et al., 2024, Li et al., 27 Aug 2025).

This suggests that cross-domain use of “DCHO” without local disciplinary context is intrinsically underdetermined. In astrochemistry, incorrect expansion changes the molecular identity; in heritage informatics, it changes the ontological status of the entity being managed; in neuroimaging, it changes the discussion from an object to a predictive architecture. For literature retrieval, database curation, and scholarly communication, the practically correct strategy is therefore not to normalize all occurrences to a single meaning, but to preserve the domain-specific expansion and, where relevant, the fully explicit symbol or acronym definition (Asensio et al., 2023, Barzaghi et al., 2024, Li et al., 27 Aug 2025).