Purpose This project aims to explore the classification method of kinase inhibitors with five-channel cell painting image data based on the deep learning model. Methods A ResNet50 transfer learning model was used as the starting point to build the deep neural network (DNN) model, where different DNN parameters were selected to make the deep learning model more suitable for the cell painting data. Two different adaptive layers (adaptive average pooling 3D and convolution 2D) were added separately before the ResNet50 transfer learning model to adapt the five-layer cell painting image to the neural network. In addition, the skimage.transform.resize function was used to compress the five-layer cell painting image. Results The proposed deep learning model demonstrates the effectiveness in all three classification experiments. The proposed model performs particularly well in classifying among control, EGFR, PIKK and CDK kinase inhibitors families. It achieves an F1-score of 0.7764 on all four targets and has a 93\% accuracy rate in the PIKK kinase inhibitors family. The adaptive average pooling 3D layer successfully adapts the five-layer images to the model, resulting in an improved effect. The training time of the model is significantly reduced to one-fortieth by compressing the image size. Conclusion The proposed model achieved convincing effectiveness in classifying families, which showed progress in building the deep learning model to classify kinase inhibitors on five-channel cell painting data. This study also proved the feasibility of directly inputting five-channel cell painting images to DNN. In addition, the speed of the model increased sharply by compressing the image size without an obvious loss of data information.

Kronisk myeloisk leukemi-celler kan överleva drogbehandling och utveckla resistens till dagens behandlingar som består av tyrosinkinasinhibitorer. Denna studie utforskar sättet på vilket K562 celler svarar till två tyrosinkinasinhibitorer, Imatinib och Bosutinib, på en transkriptionell nivå. Genom att designa studien kring ett tidsförlopp då prov för mRNA sekvensering togs vid en kontrolltidpunkt, 1, 6 och 24 timmar av behandling, samt en vecka efter 24 timmars behandling, med respektive läkemedel. K562 cellernas tillväxt var starkt hämmad av Bosutinib, även efter en veckas återhämtning från behandlingen. Imatinib-behandlade celler kunde växa nästan oförändrat både under och efter behandling. Skillnaden i inhibition av tillväxt mellan drogerna verkar även vara oberoende av dos, baserat på två testa koncentrationer för varje läkemedel: 1 µM och 3,9 µM för Imatinib, 1 µM och 0,27 µM av Bosutinib. Cellmorfologi var också ändrad av Bosutinib, då den var oförändrad vid behandling med Imatinib. Transkriptomiska analyser utfördes och gene set enrichment analysis (GSEA) användes tillsammans med over-representation analysis (ORA) för att identifiera mönster i genuttryck. Grupper av gener kopplade till nukleinsyrametabolism, RNA bearbetning och i synnerhet cellaktivering och proliferering var nedreglerade under behandling med båda läkemedlen. Efter återhämtning från behandling med Imatinib, K562 celler kunde återgå till deras ursprungliga transkriptionella profil, medan Bosutinib-behandlade celler upprätthöll långsiktig transkriptionell omprogrammering. MYC transkriptionsfaktorn var nedreglerad av både Imatinib och Bosutinib under behandlingen, och MYC kunde förknippas med en grupp av gener som var nedreglerade under läkemedelsbehandling. Resultaten i denna studie tydliggör att transkriptionell omprogrammering sker i K562 celler under behandling med TKI, och att denna omprogrammering sker på ett koordinerat sätt i grupper av gener relaterade till signaleringsvägar och viktiga cellulära processer. Hållbara förändringar i genuttryck efter Bosutinib-behandling kan länkas samman med drogens effektiva inhibition av cellernas tillväxt. / Chronic myeloid leukaemia cells are able to survive and develop resistance to current treatments consisting of tyrosine kinase inhibitors (TKIs). This study investigates the way in which K562 cells respond to two TKIs, Imatinib and Bosutinib, on a transcriptional level. Using a time course study design, mRNA sequencing (mRNA-seq) was performed on control, 1h, 6h and 24h treatment time points for each drug, as well as after one week of recovery following 24h of treatment. K562 proliferation was vastly inhibited by Bosutinib, even after one week of recovery from the 24-hour treatment period, while Imatinib-treated cells were able to proliferate almost normally during and after treatment. The difference in inhibitory effect between the two drugs seems to be dose-independent based on two tested concentrations, 1 µM and 3,9 µM Imatinib, as well as 1 µM and 0,27 µM Bosutinib. Cell morphology was also altered by Bosutinib while being unchanged during and after Imatinib treatment. Transcriptomics analysis was performed, and gene set enrichment analysis (GSEA) was used together with overrepresentation analysis (ORA) to identify patterns in gene expression. Groups of genes related to nucleic acid metabolism, RNA processing and notably regulation of cell activation and proliferation are repressed during both Imatinib and Bosutinib treatment. After recovery from Imatinib treatment, K562 cells are able to revert the transcriptional changes, while Bosutinib-treated cells sustain long-term transcriptional reprogramming. The MYC transcription factor is down-regulated by both drugs during treatment, and MYC is also linked to a collection of genes that are down-regulated during Imatinib and Bosutinib treatment. The findings from this study elucidate that transcriptional reprogramming occurs in K562 cells during TKI treatment, and that this reprogramming occurs in a concerted fashion across groups of genes related to signalling pathways and important cellular processes. Sustained changes in gene expression after Bosutinib treatment can linked to the drug’s effectiveness at inhibiting K562 cell growth.