Sentence examples for more granular annotations from inspiring English sources

What is still missing is to determine more systematically which methods are more robust or competitive for particular types of concepts or terms as well as to have more granular annotations at the level of labeling textual term evidences.

These more granular annotation types should help participants to adapt their entity recognition strategies for particularities specific to each of the chemical entity mention classes.

We therefore proposed a more granular annotation schema that covered the most important types of chemical mentions that can be found in the literature.

We also think that a more granular annotation strategy could help to improve the recognition of other entity mentions such as genes and proteins.

A more granular annotation specifically of the chemical compound mentions was proposed for the construction of the open access Chem EVAL corpus (a.k.a. SCAI corpus), a small corpus of 100 abstracts (with 1206 chemical mentions) annotated with chemical entities [35].

While the former approach is efficient in assigning large-scale higher-level GO terms, the latter provides experimentally supported, more granular GO annotations that are critical for the kinds of analyses mentioned above.

In this case the computational prediction provides more specific information than the literature-based annotation with which it is paired, potentially indicating that a more granular manual GO annotation can be made.

An example is the fatty acid metabolism role of the peroxisome, which due to the focused annotation resulted in more granular terms 'fatty acid beta oxidation using acyl CoA oxidase' being included in the annotation set thus giving better depth of information to the peroxisomal fatty acid oxidation process.

In addition to a more detailed chemical mention classification, some annotations would benefit from a more granular labeling at the level of substrings, for instance in case of hybrid chemical mentions (e.g. chemical mentions that are formed by strings belonging to different SACEM classes like SYSTEMATIC and TRIVIAL).

For example, whereas the GO abbreviation 'IEA' by default maps to ECO:0000501 'evidence used in automatic assertion', the 'IEA' evidence code in combination with GO_REF 0000002 [which references InterPro-created GO annotations (16)] maps to the more granular ECO class ECO:0000256 'match to sequence model used in automatic assertion'.

Using these new more granular terms, biologists will be able to annotate gene products in more detail.