SETH

SETH copied to clipboard

Currently, SETH does not identify synonymous protein-mutations (e.g., Asp106Asp).

Erechtheus

Some publications refer to mutations by using dbSNP ss-sequence identifier . See here for an example: https://www.sciencedirect.com/science/article/abs/pii/S0920996409003806

Erechtheus

https://github.com/biocommons/hgvs/tree/c9ee7504576d746f79135f4235a99fd6fe656852 https://github.com/mutalyzer/mutalyzer

Erechtheus

Parse dbSNP from JSON and no longer from XML

1

dbSNP changed from generating XML to JSON dunps. For building the database, we need to modify the update scripts. Data can be found at https://ftp.ncbi.nlm.nih.gov/snp/latest_release/

Erechtheus

https://github.com/ibm-aur-nlp/amia-18-mutation-corpus

Erechtheus

Improve named entity normalization using patterns

1

For more details, see the tmVar 2.0 paper (https://www.ncbi.nlm.nih.gov/research/bionlp/Tools/tmvar/), which describes a nice strategy to use patterns for named entity normalization

Erechtheus

Check tmVar 2.0 paper (https://www.ncbi.nlm.nih.gov/research/bionlp/Tools/tmvar/) They incorporate beside dbSNP also ClinVar and "hearst patterns" ftp://ftp.ncbi.nlm.nih.gov/pub/clinvar/tab_delimited/hgvs4variation.txt.gz

Erechtheus

https://www.ncbi.nlm.nih.gov/research/bionlp/Tools/tmvar/

Erechtheus

Add facility that prints mutations in BRAT format

Erechtheus

In addition to sequence-based mappings of mutations to genes occurring in the same text, we should add a few text-based mappings or checks. Oftentimes, particularly for full texts with lots...

jhkbg