PMC:2041973 / 63555-75857 JSON TXT

Annnotations TAB JSON ListView MergeView

2_test

{"project":"2_test","denotations":[{"id":"17967047-16352714-85283482","span":{"begin":143,"end":145},"obj":"16352714"},{"id":"17967047-11121071-85283483","span":{"begin":2742,"end":2744},"obj":"11121071"},{"id":"17967047-12519945-85283484","span":{"begin":5877,"end":5879},"obj":"12519945"},{"id":"T60257","span":{"begin":143,"end":145},"obj":"16352714"},{"id":"T41656","span":{"begin":2742,"end":2744},"obj":"11121071"},{"id":"T21643","span":{"begin":5877,"end":5879},"obj":"12519945"}],"text":"Materials and Methods\n\nMaintenance and differentiation of hESCs and hCNS-SCns.\nhESC line Cy203 (Cythera) was cultured as previously described [12]. To differentiate into neuroepithelial precursor cells, colonies were manually isolated from mouse embryonic fibroblasts (MEFs) and cut in small pieces. These pieces were transferred to a T75 flask with hESCs differentiation media (same hESC medium but 10% KSR and no FGF-2). Medium was changed the next day by transferring the floating hESC aggregates to a new flask. After culturing for a week, the hESC cell aggregates formed mature embroid bodies (EBs; ∼10 um round clusters with dark centers). EBs were plated on a coated 10-cm dish in hESC differentiation media. The next day, the medium was changed to DMEM/F12 supplemented with ITS and fibronectin. Medium was changed every other day for a week or until the cells formed rosette-like columnar structures that were isolated manually. These structures were then transferred to coated dishes in neural induction medium (DMEM/F12 supplemented with N2 and FGF-2) for a week. Elongated single cells were separated from leftover aggregates using non-enzymatic dissociation. After one to two passages, the cells formed a monolayer of homogeneous NPs (negative for Sox1 immunostaining). Upon confluence, cells will form neurospheres that can also be isolated from the neuroepithelial precursor cells (positive for Sox1 immunostaining). At any of these two stages, pan-neuronal differentiation can be achieved after three to four weeks. hESC line HUES6 was cultured on MEF feeders as previously described (http://www.mcb.harvard.edu/melton/hues/) or on GFR matrigel coated plates. Cells grown on matrigel were grown in MEF-conditioned medium and FGF-2 was used at 20 ng/mL instead of 10 ng/mL for cells grown on MEFs. To differentiate neuroepithelial precursors, colonies were removed by treatment with collagenase IV (Sigma) and washed three times in growth media. The pieces of colonies were resuspended in HUES growth media without FGF2 in an uncoated bacterial Petri dish to form EBs. After one week, EBs were plated on polyornathine/laminin coated plates in DMEM/F12 supplemented with N2 and FGF2. Rosette structures were manually collected and enzymatically dissociated with TryPLE (Invitrogen), plated on polyornathine/laminin coated plates, and grown in DMEM/F12 supplemented with N2 and B27-RA and 20 ng/mL FGF-2. Cells could be grown as a monolayer for up to at least ten passages. Cells were Sox1 and nestin positive and readily differentiated into neurons upon withdrawal of FGF-2. Human central nervous system stem cell line FBR1664 (StemCells) which is referred to as hCNS-SCns in the main text was cultured as previously described [23]. The cells were cultured in medium consisting of Ex Vivo 15 (BioWhittaker) medium with N2 supplement (GIBCO), FGF2 (20 ng/mL), epidermal growth factor (20 ng/mL), lymphocyte inhibitory factor (10 ng/mL), 0.2 mg/ml heparin, and 60 ug/mL N-acetylcysteine. Cultures were fed weekly and passaged at ∼two to three weeks using collagenases (Roche). The following antibodies and corresponding dilutions were utilized for the immunohistochemical analysis of marker genes in Cyt-ES and HUES6-ES: Sox2 (Chemicon, 1:500), Oct4 (Santa Cruz, 1:500), Sox1 (Chemicon, 1:500), Nestin (Pharmingen, 1:250); hCNS-SCns: Sox2 (Chemicon, 1:200), Nestin (Chemicon, 1:200).\n\nRNA preparation and array hybridization.\nTotal RNA from cells was processed as follows. Cells were lysed in 1 mL of RNA-bee (Teltest). The RNA was isolated by chloroform extraction of the aqueous phase, followed by isopropanol precipitation as per the manufacturer's instructions. The precipitated RNA was washed in 75% ethanol and eluted with DEPC-treated water. Five ug of RNA was treated with RQ1 DNAase (Promega) according to the manufacturer's instructions. One ug of total RNA for each sample was processed using the Affymetrix GeneChip Whole Transcript Sense Target Labeling Assay (Affymetrix). Ribosomal RNA was reduced with the RiboMinus Kit (Invitrogen). Target material was prepared using commercially available Affymetrix GeneChip WT cDNA Synthesis Kit, WT cDNA Amplification Kit, and WT Terminal Labeling Kit (Affymetrix) as per manufacturer's instructions. Hybridization cocktails containing ∼5 ug of fragmented and labeled DNA target were prepared and applied to GeneChip Human Exon 1.0 ST arrays. Hybridization was performed for 16 hours using the Fluidics 450 station. Arrays were scanned using the Affymetrix 3000 7G scanner and GeneChip Operating Software version 1.4 to produce .CEL intensity files.\n\nDetection of AS by RT-PCR.\ncDNAs were generated from total RNA with Superscript III reverse transcriptase (Invitrogen). PCR reactions were performed with primer pairs designed for AS targets (annealing at 58 °C and amplification for 30 or 35 cycles). PCR products were resolved on either 1.5% or 3% agarose gel in TBE. The Ethidium Bromide-stained gels were scanned with Typhoon 8600 scanner (Molecular Dynamics) for quantification. The number of true positives (TP; false negatives, FN) was computed as the number of REAP[+] (REAP[−]) exons that were validated by RT-PCR as AS. The number of true negatives (TN; or FPs) was computed as the number of REAP[−] (REAP[+]) exons that were validated by RT-PCR as constitutively spliced. The true (false) positive rate was computed as TP (FP) divided by the total number of REAP[+] exons in the experimentally validated set. The true (false) negative rate was computed as the TN (FN) divided by the total number of REAP[−] exons in the experimentally validated set. Sensitivity was computed as TP/(TP+FN) and specificity was computed as TN/(FP+TN).\n\nSequence databases.\nGenome sequences of human (hg17), dog (canFam1), rat (rn3), and mouse (mm5) were obtained from UCSC, as were the whole-genome MULTIZ alignments [80]. The lists of known human genes (knownGene containing 43,401 entries) and known isoforms (knownIsoforms containing 43,286 entries in 21,397 unique isoform clusters) with annotated exon alignments to human hg17 genomic sequence were processed as follows. Known genes that were mapped to different isoform clusters were discarded. All mRNAs aligned to hg17 that were greater than 300 bases long were clustered together with the known isoforms. Genes containing less than three exons were removed from further consideration. A total of 2.7 million spliced ESTs were mapped onto the 17,478 high-quality genes to infer AS. Exons with canonical splice signals (GT-AG, AT-AC, GC-AG) were retained, resulting in a total of 213,736 exons. Of these, 197,262 (92% of all exons) were constitutive exons, 13,934 exons (7%) had evidence of exon-skipping, 1,615 (1%) exons were mutually exclusive alternative events, 5,930 (3%) exons had alternative 3′ splice sites, 5,181 (2%) exons had alternative 5′ splice sites, and 175 (\u003c1%) exons overlapped another exon, but did not fall into the above classifications. A total of 324,139 probesets from the Affymetrix Human Exon 1.0 ST array were mapped to 208,422 human exons, representing 17,431 genes. These probesets were used to derive gene and exon-level signal estimates from the CEL files. The four-way mammalian (four-mammal) whole-genome alignment (hg17, canFam1, mm5, rn3) was extracted from the eight-way vertebrate MULTIZ alignments (hg17, panTrol1, mm5, rn3, canFam1, galGal2, fr1, danRer1) obtained from the UCSC Genome Browser. Four-way mammal alignments were extracted for all internal exons, and 400 bases of flanking intronic sequence, resulting in a total of 161,731 conserved internal exons. A total of 145,613 (90% of total) conserved internal exons were constitutive exons, 13,653 exons (8%) had evidence of exon-skipping, 1,576 exons were mutually exclusive alternative events, 5,818 exons had alternative 3′ splice sites, 5,046 exons had alternative 5′ splice sites, and 168 exons overlapped another exon.\n\nExon array analysis.\nThe Affymetrix Power Tools (APT) suite of programs was obtained from http://www.affymetrix.com/support/developer/powertools/index.affx. Exon (probeset) and gene-level signal estimates were derived from the CEL files by RMA–sketch normalization as a method in the apt-probeset-summarize program. To determine if the signal intensity for a given probeset is above the expected level of background noise, we utilized the DABG (detection above background) quantification method available in the apt-probeset-summarize program as part of Affymetrix Power Tools (APT). Briefly, DABG compared the signal for each probe to a background distribution of signals from anti-genomic probes with the same GC content. The DABG algorithm generated a p-value representing the probability that the signal intensity of a given probe was part of the background distribution. We considered a probeset with a DABG p-value lower than 0.05 as detected above background. The statistic thCNS-SCns,ESC = (μhCNS-SCns − μESC) / sqrt (((nhCNS-SCns − 1)σ2 hCNS-SCns + (nESC − 1)σ2 ESC)(nhCNS-SCns + nESC)) / ((nhCNS-SCnsnESC) (nhCNS-SCns + nESC − 2))), where nhCNS-SCns and nESC were the number of replicates, μhCNS-SCns and μESC were the mean, and σ2 hCNS-SCns and σ2 ESC were the variances of the expression values for the two datasets used to represent the differential enrichment of a gene using gene-level estimates in hCNS-SCns relative to hESCs. Multiple hypothesis testing was corrected by controlling for the false discovery rate (Benjamini-Hochberg).\n\nAS detection by REAP.\nThe log2 signal estimate xij for probeset i in cell-type j had to satisfy two conditions, otherwise the probeset was discarded: (i) 2 \u003c xij \u003c 10,000 for all conditions/cell-types j; and (ii) DABG p-value \u003c 0.05 for all replicates in at least one condition/cell-type j. A gene had to have five probesets that satisfied the two conditions above in order to be considered for robust regression analysis. After generating the points (as described in the Results section), we utilized the robust regression method rlm in R-package “MASS” (version 6.1–2) with M-estimation and a maximum iteration setting of 30 to estimate the linear function yi = αxi + β. For each probeset, we computed the error term ei,, which was the difference between the actual value yi and the estimated value ξi, from the estimated function ξi = Axi + B, where A and B were estimates of α and β. The error term variance was estimated by se 2 = Σei 2/(n − p), which was used to estimate the variance of the predicted value, sξi 2 = se 2(n−1 + (xi − μx)2 / sx 2(n − 1)). Here, n referred to the number of points (generated for each gene), and p referred to the number of independent variables (p = 2 in our method); and μx = Σxi 2/n; sx 2 = n−1 Σ(xi − μx)2. Following Belsley et al. [81], we defined the leverage hi of the ith point as hi = n−1 + (xi − μx)2 / sx 2(n − 1). Here we considered a point to have high leverage if hi \u003e 3p/n. Next, we calculated the covariance ratio, covi = (si 2/sr 2)p/(1 − hi), which is the ratio of the determinant of the covariance matrix after deleting the ith observation to the determinant of the covariance matrix with the entire sample. We considered a point to have high influence if |covi − 1| \u003e 3p/n. Lastly, we computed the studentized residuals, rstudenti = ei / (s(i) 2 (1 − hi)0.5), where s(i) 2 = (n-p)se 2 / (n-p-1) – ei 2 / (n-p-1)(1 − hi), the error term variance after deleting the ith point. As rstudenti was distributed as Student's t-distribution with n-p-1 degrees of freedom, each rstudenti value was associated with a p-value. We considered a point to be an “outlier” if p \u003c 0.01.\n\nIdentification of motifs.\nThe enrichment score of a sequence element of length k (k-mer) in one set of sequences (set 1) versus another set of sequences (set 2) was represented by the nonparametric χ2 statistic with Yates correction, computed from the two by two contingency table, T (T11: number of occurrences of the element in set 1; T12: number of occurrences of all other elements of similar length in set 1; T21: number of occurrences of element in set 2; T22: number of occurrences of all other elements of similar length in set 2. All elements had to be greater than 5. To correct for multiple hypothesis testing, p-values were multiplied by the total number of comparisons.\n"}

CellFinder

PMC:2041973 / 63555-75857 JSONTXT

Annnotations TAB JSON ListView MergeView

2_test

CellFinder

PMC:2041973 / 63555-75857 JSON TXT