[HTML][HTML] An expanded evaluation of protein function prediction methods shows an improvement in accuracy

…, C Dessimoz, T Dogan, K Hakala, S Kaewphan… - Genome biology, 2016 - Springer
Background A major bottleneck in our understanding of the molecular underpinnings of life
is the assignment of function to proteins. While molecular experiments provide the most …

[HTML][HTML] The CAFA challenge reports improved protein function prediction and new functional annotations for hundreds of genes through experimental screens

…, A Benso, K Hakala, F Ginter, F Mehryary, S Kaewphan… - Genome biology, 2019 - Springer
Background The Critical Assessment of Functional Annotation (CAFA) is an ongoing, global,
community-driven effort to evaluate and improve the computational annotation of protein …

[PDF][PDF] UTurku: drug named entity recognition and drug-drug interaction extraction using SVM classification and domain knowledge

J Björne, S Kaewphan, T Salakoski - Second Joint Conference on …, 2013 - aclanthology.org
… External data is added to the head token features, from where it is combined into more
complex features. One example is generated for each token in the sentence, and these are …

[HTML][HTML] Lysophosphatidic acid and sphingosine-1-phosphate promote morphogenesis and block invasion of prostate cancer cells in three-dimensional organotypic …

…, P Kohonen, P Kovanen, A Happonen, S Kaewphan… - Oncogene, 2012 - nature.com
Normal prostate and some malignant prostate cancer (PrCa) cell lines undergo acinar
differentiation and form spheroids in three-dimensional (3-D) organotypic culture. Acini formed by …

Neural network and random forest models in protein function prediction

K Hakala, S Kaewphan, J Björne… - IEEE/ACM …, 2020 - ieeexplore.ieee.org
… Suwisa Kaewphan received the MSc degree in bioinformatics from the University of Turku
and is currently working toward the PhD degree in computer science at the Department of …

[PDF][PDF] Syntactic analyses and named entity recognition for PubMed and PubMed Central—up-to-the-minute

K Hakala, S Kaewphan, T Salakoski… - Proceedings of the 15th …, 2016 - aclanthology.org
… The aim of this work is to support the future development of largescale text mining resources
… our resource this is done weekly — the same interval the official PMCOA dataset is updated. …

Cell line name recognition in support of the identification of synthetic lethality in cancer from text

S Kaewphan, S Van Landeghem, T Ohta… - …, 2016 - academic.oup.com
… Results: We find that the best performance is achieved using NERsuite, a machine learning
system based on Conditional Random Fields, trained on the Gellus corpus and supported …

Wide-scope biomedical named entity recognition and normalization with CRFs, fuzzy matching and character level modeling

S Kaewphan, K Hakala, N Miekka, T Salakoski… - Database, 2018 - academic.oup.com
… Providing this information is crucial as the CNN-BiLSTM-CRF model is not given any word
level information and is unable to achieve good performance based solely on the characters. …

[PDF][PDF] UTU: Disease mention recognition and normalization with CRFs and vector space representations

S Kaewphan, K Hakala, F Ginter - Proceedings of the 8th …, 2014 - aclanthology.org
In this paper we present our system participating in the SemEval-2014 Task 7 in both subtasks
A and B, aiming at recognizing and normalizing disease and symptom mentions from …

End-to-end system for bacteria habitat extraction

F Mehryary, K Hakala, S Kaewphan, J Björne… - BioNLP …, 2017 - aclanthology.org
… a token is present in the ontology or not, a similarity score ranging from 0 to 1 is assigned for
each … According to our evaluation this is not beneficial for the BACTERIA entities and is thus …