A Computational Framework for Heparan Sulfate Sequencing Using High-resolution Tandem Mass Spectra
Bibliographic record
Abstract
Heparan sulfate (HS) is a linear polysaccharide expressed on cell surfaces, in extracellular matrices and cellular granules in metazoan cells. Through non-covalent binding to growth factors, morphogens, chemokines, and other protein families, HS is involved in all multicellular physiological activities. Its biological activities depend on the fine structures of its protein-binding domains, the determination of which remains a daunting task. Methods have advanced to the point that mass spectra with information-rich product ions may be produced on purified HS saccharides. However, the interpretation of these complex product ion patterns has emerged as the bottleneck to the dissemination of these HS sequencing methods. To solve this problem, we designed HS-SEQ, the first comprehensive algorithm for HS de novo sequencing using high-resolution tandem mass spectra. We tested HS-SEQ using negative electron transfer dissociation (NETD) tandem mass spectra generated from a set of pure synthetic saccharide standards with diverse sulfation patterns. The results showed that HS-SEQ rapidly and accurately determined the correct HS structures from large candidate pools. Heparan sulfate (HS) is a linear polysaccharide expressed on cell surfaces, in extracellular matrices and cellular granules in metazoan cells. Through non-covalent binding to growth factors, morphogens, chemokines, and other protein families, HS is involved in all multicellular physiological activities. Its biological activities depend on the fine structures of its protein-binding domains, the determination of which remains a daunting task. Methods have advanced to the point that mass spectra with information-rich product ions may be produced on purified HS saccharides. However, the interpretation of these complex product ion patterns has emerged as the bottleneck to the dissemination of these HS sequencing methods. To solve this problem, we designed HS-SEQ, the first comprehensive algorithm for HS de novo sequencing using high-resolution tandem mass spectra. We tested HS-SEQ using negative electron transfer dissociation (NETD) tandem mass spectra generated from a set of pure synthetic saccharide standards with diverse sulfation patterns. The results showed that HS-SEQ rapidly and accurately determined the correct HS structures from large candidate pools. Heparan sulfate (HS) 1The abbreviations used are: HS, heparan sulfate; Ac, acetate; AT, antithrombin; HSPG, heparan sulfate proteoglycan; ECM, extracellular matrix; FGF, fibroblast growth factor; VEGF, vascular endothelial growth factor; GDNF, glial cell line-derived neurotrophic factor; GAG, glycosaminoglycan; GP, golden pair; EDD, electron detachment dissociation; CID, collision-induced dissociation; NETD, negative electron transfer dissociation; RE, reducing end; NRE, non-reducing end; FTICR, Fourier transform ion cyclotron resonance; S/N, signal-to-noise ratio; Ac, acetate; GUI, graphical user interface; MVC, model-view-controller; Me, methyl; AnMan, 2,5-anhydro-D-mannose; PNP, 4-nitrophenol. 1The abbreviations used are: HS, heparan sulfate; Ac, acetate; AT, antithrombin; HSPG, heparan sulfate proteoglycan; ECM, extracellular matrix; FGF, fibroblast growth factor; VEGF, vascular endothelial growth factor; GDNF, glial cell line-derived neurotrophic factor; GAG, glycosaminoglycan; GP, golden pair; EDD, electron detachment dissociation; CID, collision-induced dissociation; NETD, negative electron transfer dissociation; RE, reducing end; NRE, non-reducing end; FTICR, Fourier transform ion cyclotron resonance; S/N, signal-to-noise ratio; Ac, acetate; GUI, graphical user interface; MVC, model-view-controller; Me, methyl; AnMan, 2,5-anhydro-D-mannose; PNP, 4-nitrophenol. is required for all aspects of physiology in metazoan cells and tissues (1.Bishop J.R. Schuksz M. Esko J.D. Heparan sulphate proteoglycans fine-tune mammalian physiology.Nature. 2007; 446: 1030-1037Crossref PubMed Scopus (1265) Google Scholar, 2.Parish C.R. The role of heparan sulphate in inflammation.Nat. Rev. Immunol. 2006; 6: 633-643Crossref PubMed Scopus (385) Google Scholar, 3.Ori A. Wilkinson M. Fernig D. The heparanome and regulation of cell function: structures, functions, and challenges.Front. Biosci. J. Virtual Libr. 2008; 13: 4309Crossref PubMed Scopus (130) Google Scholar). As a subclass of glycosaminoglycans (GAGs), HS consists of repeating disaccharide units [-4-(GlcA-β/IdoA-α)-1,4-GlcNAc(NS)-α-], where GlcA/IdoA may undergo 2-O-sulfation, and GlcNAc may undergo N-deacetylation (free -NH2), N-sulfation, 6-O-sulfation, and in rare cases, 3-O-sulfation (1.Bishop J.R. Schuksz M. Esko J.D. Heparan sulphate proteoglycans fine-tune mammalian physiology.Nature. 2007; 446: 1030-1037Crossref PubMed Scopus (1265) Google Scholar, 2.Parish C.R. The role of heparan sulphate in inflammation.Nat. Rev. Immunol. 2006; 6: 633-643Crossref PubMed Scopus (385) Google Scholar). HS chains bind covalently to core proteins via a tetrasaccharide linker to a serine residue, and form heparan sulfate proteoglycans (HSPGs). The HS chains on HSPG bind non-covalently to growth factors, morphogens, chemokines, cytokines, and other molecules, and regulate biological activities including cell signaling, cell migration, and remodeling of the extracellular matrix (ECM) (1.Bishop J.R. Schuksz M. Esko J.D. Heparan sulphate proteoglycans fine-tune mammalian physiology.Nature. 2007; 446: 1030-1037Crossref PubMed Scopus (1265) Google Scholar, 3.Ori A. Wilkinson M. Fernig D. The heparanome and regulation of cell function: structures, functions, and challenges.Front. Biosci. J. Virtual Libr. 2008; 13: 4309Crossref PubMed Scopus (130) Google Scholar, 4.Bülow H.E. Hobert O. The molecular diversity of glycosaminoglycans shapes animal development.Annu. Rev. Cell Dev. Biol. 2006; 22: 375-407Crossref PubMed Scopus (271) Google Scholar, 5.Couchman J.R. Transmembrane signaling proteoglycans.Annu. Rev. Cell Dev. Biol. 2010; 26: 89-114Crossref PubMed Scopus (297) Google Scholar). Spatial and temporal regulation of the activities of HS biosynthetic enzymes results in mature HS structure that varies according to the biological context. Mature HS chains consist of domains of high, low, and intermediate sulfation respectively, the organization of which varies according to cell type and developmental state. Moreover, variation of chain length and modification patterns associated with different biological sources adds complexity to the HS chains (6.Esko J.D. Lindahl U. Molecular diversity of heparan sulfate.J. Clin. Invest. 2001; 108: 169-173Crossref PubMed Scopus (787) Google Scholar). The diversity of HS structures and different levels of binding specificity with multiple protein ligands suggests a physiological mechanism of cellular responses to growth factor stimuli. A well-studied case is the highly sulfated pentasaccharide sequence within heparin that binds antithrombin (AT)-III specifically and regulates blood coagulation process. This pentasaccharide motif contains eight sulfate substituents, including a 3-O-sulfate group, each of which is required for high affinity binding (7.Lindahl U. Bäckström G. Thunberg L. Leder I.G. Evidence for a 3-O-sulfated D-glucosamine residue in the antithrombin-binding sequence of heparin.Proc. Natl. Acad. Sci. U. S. A. 1980; 77: 6551-6555Crossref PubMed Scopus (418) Google Scholar, 8.Atha D.H. Lormeau J.C. Petitou M. Rosenberg R.D. Choay J. Contribution of monosaccharide residues in heparin binding to antithrombin III.Biochemistry. 1985; 24: 6723-6729Crossref PubMed Scopus (189) Google Scholar, 9.Petitou M. van Boeckel C.A.A. A synthetic antithrombin iii binding pentasaccharide is now a drug! What comes next?.Angew. Chem. Int. Ed. 2004; 43: 3118-3133Crossref PubMed Scopus (415) Google Scholar). Moreover, studies on HS-binding proteins, including fibroblast growth factors (FGF) (10.Schlessinger J. Plotnikov A.N. Ibrahimi O.A. Eliseenkova A.V. Yeh B.K. Yayon A. Linhardt R.J. Mohammadi M. Crystal structure of a ternary FGF-FGFR-heparin complex reveals a dual role for heparin in FGFR binding and dimerization.Mol. Cell. 2000; 6: 743-750Abstract Full Text Full Text PDF PubMed Scopus (961) Google Scholar, 11.Zhang F. Zhang Z. Lin X. Beenken A. Eliseenkova A.V. Mohammadi M. Linhardt R.J. Compositional analysis of heparin/heparan sulfate interacting with fibroblast growth factor·fibroblast growth factor receptor complexes.Biochemistry. 2009; 48: 8379-8386Crossref PubMed Scopus (58) Google Scholar, 12.Naimy H. Buczek-Thomas J.A. Nugent M.A. Leymarie N. Zaia J. Highly sulfated nonreducing end-derived heparan sulfate domains bind fibroblast growth factor-2 with high affinity and are enriched in biologically active Biol. Chem. Full Text Full Text PDF PubMed Scopus Google Scholar, M. and of multiple heparan sulfate proteoglycans and fibroblast growth factor Biol. Chem. 2008; Full Text Full Text PDF PubMed Scopus Google proteoglycans regulate responses PubMed Google vascular endothelial growth factor S. D. L. G. the of with and cell associated Biol. Chem. Full Text PDF PubMed Google Scholar, Nugent M.A. regulates vascular endothelial growth factor binding to a mechanism for of extracellular matrix and Biol. Chem. 2004; Full Text Full Text PDF PubMed Scopus Google glial cell line-derived neurotrophic factor The binding of glial cell line-derived neurotrophic factor to heparin and heparan of and on its with its 13: PubMed Scopus Google M. H. M. Cell on cell in fine structure and binding of its heparan sulfate Biol. Chem. Full Text PDF PubMed Google and X. O. M. Lindahl U. the sulfation of cell heparan sulfate proteoglycans to Cell Biol. PubMed Scopus Google the of sulfation patterns of HS chains in the binding specificity HS, growth factors, and growth factor and the of and F. D. H. A synthetic heparan to a high of Biol. Full Text Full Text PDF PubMed Scopus Google Scholar, heparan sulfate proteoglycans in 2008; PubMed Scopus Google Scholar, N. A. biological functions, and of heparan 2007; 13: PubMed Scopus Google Scholar, U. Heparan for 2007; PubMed Scopus Google Scholar). this for sulfation patterns on HS chains are in high The of the has the of HS sequencing X. X. Leymarie N. H. Zaia J. of heparan sulfate via Chem. PubMed Scopus Google Scholar, L. M. Linhardt R.J. PubMed Scopus Google Scholar, X. Lin electron dissociation of Chem. PubMed Scopus Google Scholar, X. Lin Chem. PubMed Scopus Google Scholar). in of HS with sulfation have in with S. M. H. J. Linhardt R.J. J. of molecular PubMed Scopus Google Scholar). tandem mass dissociation are now of sulfation patterns of M. Linhardt R.J. The has a Chem. Biol. PubMed Scopus Google Scholar, X. Linhardt R.J. L. analysis and in the Biol. Chem. Full Text Full Text PDF PubMed Scopus Google Scholar). electron detachment dissociation L. Linhardt R.J. from in using electron detachment Chem. 2007; PubMed Scopus (130) Google and negative electron transfer dissociation (NETD) Linhardt R.J. electron transfer dissociation of Chem. 2010; PubMed Scopus Google Scholar, Z. M. S. A. Linhardt R.J. electron transfer dissociation Fourier transform mass of J. PubMed Scopus Google spectra that and the sulfation is to of sulfate in collision-induced dissociation tandem mass spectra HS X. H. Zaia J. mass of heparan sulfate negative sulfate patterns and modification for of product ion PubMed Scopus Google with L. M. Linhardt R.J. mass of a synthetic heparin using collision-induced Chem. PubMed Scopus Google Scholar, J. mass of sulfated Chem. PubMed Scopus Google Scholar). However, all tandem mass spectra of highly sulfated HS dissociation with of sulfate This to a of product ions with sulfate the interpretation of which the dissemination of these is to to the of these which the of the in and in of HS fine structure on physiological activities. the heparin sequencing using and a J.A. sequencing using and with a sequencing Chem. 77: PubMed Scopus Google Scholar). on analysis that tandem mass spectra of disaccharide units produced from the saccharide in from the Methods have advanced that and now determination of sulfation and from tandem mass spectra A to the fine structure and organization of HS sequence using from and biosynthetic J.A. Buczek-Thomas J.A. Nugent M.A. A for the organization of 2010; PubMed Scopus Google Scholar). This produced HS chain and is for the of candidate to the of sulfate for a The A. H. A. a for the of mass spectra of 2008; PubMed Scopus Google Scholar, A. A. for the determination of glycosaminoglycans mass Chem. 2008; PubMed Scopus Google Scholar, D. A. A. The and and Chem. PubMed Scopus Google to the of in tandem mass spectra of in the of sulfated on the sequence is sequencing HS and other from tandem mass spectra. The mass a set of candidate of biosynthetic The in the tandem mass spectra candidate to with the of the on a is required for candidate tandem sequencing are and de novo novo sequencing J. for the de novo sequencing of from tandem mass Rev. PubMed Scopus Google on a of structures, and is for HS type of de novo sequencing is on the algorithm for sequencing mass Scopus Google Scholar, J.A. via de novo sequencing tandem mass PubMed Scopus Google Scholar, novo sequencing via tandem mass Biol. 6: PubMed Scopus Google Scholar, J.A. and of de novo sequencing tandem mass Chem. 2001; PubMed Scopus Google Scholar, A for the of and ions in tandem mass PubMed Scopus Google Scholar, A de novo sequencing for using high-resolution tandem mass 2010; PubMed Scopus Google which the sequencing the product ion from to the this mass the HS chains consist of repeating and residues with and sulfate as This that product ions may to monosaccharide sulfate the of product ions is product ions may mass A J. for the de novo sequencing of from tandem mass Rev. PubMed Scopus Google as Zhang M. A. G. for de novo sequencing tandem mass PubMed Scopus Google product and candidate with the This on the of ion and with may its HS, the of the sequence is a that on the and of a with Ac, to a with may the the and on with of the may be to the sequence from candidate via sequence the with may be product ions of product ion The HS sequencing be as the to of the The mass the HS the of monosaccharide and the of monosaccharide residues the sequence of monosaccharide residues and the sequence of candidate the this we the HS sequence as the to of a set of candidate We in to the We modification to the of which are on the sequence we the as of the is of its a candidate modification a on a monosaccharide residue where the modification may and modification for the of the modification The is a of modification that the of are all the candidate modification modification type to a modification and the of the the modification in the The of candidate modification and modification as the for the of a we each interpretation of a ion and modification as in the candidate modification and modification this each a of the HS sequencing problem, which to the modification to Moreover, a monosaccharide residue of the reducing non-reducing and each modification we to the the the and the the of candidate modification of to the HS the is to the mass of to and the of to the modification of the sequence a mass to multiple and different the sequence the is all of the candidate modification modification As we have all the and in HS sequencing be in of candidate modification and modification on the we a de novo sequencing HS-SEQ, to the of HS As in the sequencing in HS-SEQ of the of the of the and the of the of sulfate a sulfate on each monosaccharide residue and sulfation within this to and to the modification HS-SEQ and the the The of each the modification The modification be to a of candidate of We tested HS-SEQ using synthetic HS saccharide standards a of chain modification sulfation and ion The results showed that HS-SEQ accurately the correct HS for of of the tested tandem mass and the correct for the each of the tandem spectra with a different the sequence as Moreover, the for the correct HS structures from and the required for sequencing a the of a HS sequencing The using the and is for The and a ion and The that the of and A of the is and be in To the we is the first to the of HS We showed in the of and sulfate in tandem mass spectra of HS and mass spectra of the synthetic HS on a transform ion cyclotron mass in HS standards in to a of and the mass using The spectra in the negative ion and the to ions using a and in a cell tandem generated in a in the of dissociation using a of to and a of to with each tandem mass to for The using tandem from the spectra using with the and the signal-to-noise set to of the the and for a and to the of The mass and to mass with a mass The from the of ion Ac, product ions A for in spectra of J. Scopus Google including of and ions generated from multiple of the HS mass transfer and all of on the product ions We that for sulfation may and and for sulfation may the has that the residues in different patterns using L. Linhardt R.J. from in using electron detachment Chem. 2007; PubMed Scopus (130) Google the has as for from the spectra. HS-SEQ to from A of a of product ion type and and modification for and sulfate with the product ion type and the of other each with mass in the of a of mass transfer and sulfate that of and ions and ions with with of the different each the with the of sulfate to the the sulfate in HS-SEQ the the mass in of The of set to be from to The sulfate of the As a each of a of the that the of of sulfate for This of the of modification in sequence a to be to a to be to the candidate different and be from the mass involved the of sulfate of sulfate dissociation may to of the the with the of sulfate within a the of sulfate in HS-SEQ to the of be in each modification the algorithm to the modification for the used for the of each modification the of a as is via as and is the of that to the mass is the of a in is the associated with the type and the of of set to for of type set to for set to which of that the of the on with of modification and modification of a mass the modification and modification be the of the via for each modification type the set of candidate modification of and the of of the is a of and is that is a of that is for a to have and are of the with to the sulfation is as the of and the modification of the the set and is the modification of the is a of and the are to each As in the with the candidate modification and modification of the and the other a with modification and modification of and in the The the for and in the and the on The of on the the and the with the and The of where is the of a and is a within The of the of is its as as the and the of the modification of to the from different regulates the of the on the The this is that the the modification of is to the the the is to the modification This is for the sulfate we to that are to the sulfation and in are where is the of the is the of the of is the of the of is the of the of is the modification and is the of candidate modification of has in the is the from a of for golden the from the for the and is the from the for the the that candidate the and its to the each with the that for sulfation with the and modification in that the with the of sulfate in each This is for the candidate sulfation and The of involved the with in the as and the with of the as A of and the of is the of the is the of the of and the of in the of with of The in a and the of a with the of its in the the modification to on all modification the The of a the to of a of the modification the of the the for the of for the from the of modification the of the the to the on the modification to the modification on the modification of the the of modification of on the modification The modification for is the of the the modification is the and is the modification for where is the of the is the of the of is the modification is the of candidate modification and is the modification the candidate modification the and of the is the S. the modification is the of modification is the of the of is the of the of the the modification in the for the the and its the modification on each candidate modification modification for and its the modification on each candidate As the the modification of with the candidate modification and modification The first the candidate on that in to the of sulfation the of product ions the sulfate within each product ions to be associated with sulfate As a HS-SEQ the sulfate within the residue the of sulfate on each residue to the of sulfation within residues on the biosynthetic of HS H.E. Hobert O. The molecular diversity of glycosaminoglycans shapes animal development.Annu. Rev. Cell Dev. Biol. 2006; 22: 375-407Crossref PubMed Scopus (271) Google Scholar, J.D. of of binding in heparan Rev. PubMed Scopus Google Scholar). The of HS-SEQ using spectra from each all on the of the and the for each candidate sequence candidate sequence modification with where and and The for a candidate sequence as the the modification the candidate sequence and the modification HS-SEQ, which is the set of modification is the set of candidate modification for modification type is the of modification from HS-SEQ on for modification type and is the modification candidate sequence on for modification type each the and of the of the sequence in the of all candidate The sequence to have a and high from of the using each the of the sequence in the sequence and sequence using HS-SEQ in a The of HS-SEQ with other methods. The first the A. H. A. a for the of mass spectra of 2008; PubMed Scopus Google Scholar, D. A. A. The and and Chem. PubMed Scopus Google which we as a candidate sequence the of mass of product ions that the sequence have the and all other a of the mass The tested the J. mass of sulfated Chem. PubMed Scopus Google which we as the of for a candidate we that a sequence have the of golden from biological a for sulfation within a the of a of all which we as and HS-SEQ with in the of sulfate for each residue to have the HS-SEQ for a sequence the residue is the set of modification is the of monosaccharide residues for modification type is the modification from HS-SEQ on residue for modification type and is the modification the candidate sequence on residue for modification type that the is to variation of sulfation within each HS-SEQ sulfation on in a residue as and and a candidate sequence the sulfation on the as and The of the the as the a of type to sulfate the on the the candidate as as the of the candidate sequence for candidate sequence in a to sulfate group, and a to sulfate the a sulfate and a to sulfate a sequence with sulfation and on a residue of as sequence with sulfation and on the residue, other residues of the the sulfation The to to for each candidate The and of the The the and the of candidate sequence for the and of all The results are in and the modification HS-SEQ are in We the of the using all A high The results that the the the and for sequence and and The of HS-SEQ to the in with the that for sequence and and and sequence The for and and and and The a of the of a sequence from the HS-SEQ for sequence and and The for and produced results for of of golden The the in the a in the This in HS tandem spectra to the of candidate The results showed that HS-SEQ for sequence sequence sequence sequence and We the linear and The results showed that the linear for HS-SEQ with for the and methods. for HS-SEQ to have a a for a sequencing algorithm to of the HS-SEQ to with with the methods. This suggests that factors other the may the of HS-SEQ The for HS-SEQ a with other which suggests that the from HS-SEQ the all showed linear on the the and the for the for the and for HS-SEQ the for HS-SEQ the for the methods. This that HS-SEQ in the sequence from the in the case of large The of HS-SEQ highly for HS of the modification from HS-SEQ of sulfate with all dissociation including X. Zaia J. Lin novo sequencing of heparan sulfate Chem. PubMed Scopus Google Scholar). This to for the product ions with sulfate the and HS-SEQ required the of product ions with sulfate the of sulfate on each residue in the sulfation generated HS-SEQ, each residue to all sulfate this the sulfation within we HS-SEQ to in the We the of all As the HS-SEQ showed in the and its the in the for its in the As in HS-SEQ the correct structure as for of and all in multiple the of each sequence from multiple spectra to a We that of the spectra of the sequence the sequence as the sequence The results showed that HS-SEQ all the correct structure for of the and and for sequence of the The showed results in the and as the and the the linear for all HS-SEQ, from to for the from to for the from to the of HS-SEQ and the other The linear the and the for all for HS-SEQ for and for the for HS-SEQ the This suggests that as the HS-SEQ to the the the of the and the as for of the tested spectra. However, the all in the the of are in and The of for all is in To HS-SEQ the the and which and HS sequencing using tandem mass The of HS-SEQ be as the A of the user in the in this the modification of the the the and the to modification and the the of the The HS-SEQ algorithm used a to the sequence of candidate required in the process. The and required a to the modification from where the candidate sequence be of and tandem mass spectra from the sequence modification the of modification each modification This is to results from different and dissociation and of the sequence and required for HS-SEQ to The first the of tandem mass spectra where ions with sulfate and HS-SEQ, of product ions with sulfate for structure The required that a of product ions be The results for HS the reducing to the with may the and to sequencing in which case be a with and the of is that a mechanism to the of sulfation from the of sulfate from may the the a may sulfate its The may from multiple as sulfate ion and of HS-SEQ the with in to the This be for The and of the in may the of HS-SEQ and to of the HS from may from the and and the of de novo sequencing and we designed a for HS de novo We that the to other J. using mass Cell. Full Text Full Text PDF PubMed Scopus Google all and The used in may be to the of high-resolution tandem sequencing for other complex molecules, the sequencing and have We for in the with
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".