Computational protocol: Determination of genes and microRNAs involved in the resistance to fludarabine in vivo in chronic lymphocytic leukemia

Similar protocols

Protocol publication

[…] Microarray experiments were performed with samples from CLL patients sensitive (CLL-1, -2, -4, -5) and resistant (CLL-3 and -6) to fludarabine. As the treatment of patients can be assimilated to a time course experiment, the commonly used two-channel microarray experiment loop design was chosen []. Each sample was hybridized to two different samples in two different dye orientations (T0 vs T1, T1 vs T2, T2 vs T9, T9 vs T0) in quadruplicates. Briefly, 1 μg of total RNA from B cells of patients CLL-1 to -6 only was amplified and labeled with the Amino Allyl MessageAmp™ II aRNA Amplification Kit (Ambion, UK). The coupling reaction was performed following the protocol of the manufacturer using Alexa Fluor® Reactive Dye Decapacks (AF555 and 647, Molecular Probes). Alexa Fluor®-aRNA were hybridized for 16 h on cDNA microarrays (Operon human version2 oligonucleotide library, University Medical Center Utrecht, The Netherlands). Microarrays were scanned at 10-μm resolution with two laser channels (532 and 635 nm) using a GenePix® 4000B scanner and GenePix® Pro 6.0 (Molecular Devices Corporation, CA) at variable photomultiplier tube (PMT) gains (< 1% saturated spots).Microarray data were deposited in the EMBL-EBI ArrayExpress public repository (Accession number E-MTAB-70). The image analysis and spot selection were performed with MAIA 2.7 (Institut Curie, France) as previously described []. Acuity® 4.0 (Molecular Devices, CA) was used for data processing (dye-swap, Lowess normalization), warehousing and visualization. A few microarrays were excluded from analysis based on the results of Pearson cross-correlation coefficient (data not shown). Data were subjected to statistical examination with the SAM add-on for Microsoft Excel (Significance Analysis of Microarrays, Stanford, CA; using centered median, 100 permutations). Missing values were imputed using the k-nearest neighbor method (k = 10). Delta values were adjusted to obtain gene lists with a false discovery rate (FDR) ≤ 1% for one-class analysis and ≤ 5% for two-class analysis. Genes identified by SAM two-class analysis as regulated in CLL B cells after 24 h of treatment with fludarabine in vivo were classified by enriched gene sets with the Gene Set Enrichment Analysis algorithm (GSEA 2.05, database Dec 2009; Broad Institute, MA). Differentially regulated genes were clustered as molecular interaction networks with Ingenuity Pathway Analysis (IPA 8.0, database Nov 2009; Ingenuity Systems, CA). […]

Pipeline specifications

Software tools GenePix Pro, IPA
Databases ArrayExpress
Application Gene expression microarray analysis
Organisms Homo sapiens
Diseases Leukemia, Lymphocytic, Chronic, B-Cell