Back

Trending Drugs Combination to Target Leukemia associated Proteins/Genes: using Graph Neural Networks under the RAIN Protocol

Boush, M.; Kiaei, A. A.; Mahboubi, H.

2023-08-21 geriatric medicine
10.1101/2023.08.17.23294228 medRxiv
Show abstract

BackgroundLeukemia, a cancer impacting blood-forming tissues such as bone marrow and the lymphatic system, presents in various forms, affecting children and adults differently. The therapeutic approach is complex and depends on the specific leukemia type. Effective management is crucial as it disrupts normal blood cell production, increasing infection susceptibility. Treatments like chemotherapy can further weaken immunity. Thus, a patients healthcare plan should focus on comfort, reducing chemotherapy side effects, protecting veins, addressing complications, and offering educational and emotional support. MethodThis article reviews studies on the combined use of drugs for treating leukemia. Employing a mix of medicines might decrease the chances of tumor resistance. Starting multiple drugs concurrently allows for immediate application during disease onset, avoiding delays. Initial chemotherapy uses a drug combination to eliminate maximum leukemia cells and restore normal blood counts. Afterwards, intensification chemotherapy targets any residual, undetectable leukemia cells in the blood or bone marrow. To recommend a drug combination to treat/manage Leukemia, under first step of RAIN protocol, we have searched articles including related trend drugs using Natural Language Processing. In the second step, we have employed Graph Neural Network to pass information between these trending drugs and genes that act as potential targets for Leukemia. ResultAs a result, the Graph Neural network recommends combining Tretinoin, Asparaginase, and Cytarabine. The network meta-analysis confirmed the effectiveness of these drugs on associated genes. ConclusionThe p-value between leukemia and the scenario that includes combinations of the mentioned drugs is almost zero, indicating an improvement in leukemia treatment. Reviews of clinical trials on these medications support this claim. HighlightsO_LICombined drugs that make p-value between Leukemia and target proteins/genes close to 1 C_LIO_LIUsing Graph Neural network to recommend drug combination C_LIO_LIA Network meta-analysis to measure the comparative efficacy C_LIO_LIConsidered drug interactions C_LI O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=96 SRC="FIGDIR/small/23294228v1_ufig1.gif" ALT="Figure 1"> View larger version (37K): org.highwire.dtl.DTLVardef@1347a06org.highwire.dtl.DTLVardef@374783org.highwire.dtl.DTLVardef@138022forg.highwire.dtl.DTLVardef@179be2_HPS_FORMAT_FIGEXP M_FIG C_FIG

Matching journals

The top 4 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.