Back

Comparing machine learning models for predicting mutation status in Acute Myeloid Leukemia patients using RNA-seq data

Silva, R.; Riedel, C.; Reboul, J.; Guibert, B.; Ruffle, F.; Gallopin, M.; Gilbert, N.; Boureux, A.; Commes, T.

2024-11-14 bioinformatics
10.1101/2024.11.13.623391 bioRxiv
Show abstract

Acute Myeloid Leukemia (AML) is a highly heterogeneous disease. The current AML classifications are based mainly on molecular markers, including cytogenetics features, fusion genes, and the presence or absence of mutations. In this study, we investigated mutation status in AML patients through RNA-seq data in link with differential gene expression. We applied seven machine learning algorithms to identify the presence or absence of NPM1, IDH1/IDH2, and FLT3-ITD mutations, reaching 95%, 93%, and 87% accuracy, respectively. In each case, the best performing models were complex models, suggesting highly complex biological processes at work behind AML.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.