The transcriptional landscape of the mammalian genome

This study describes comprehensive polling of transcription start and termination sites and analysis of previously unidentified full-length complementary DNAs derived from the mouse genome. We identify the 5' and 3' boundaries of 181,047 transcripts with extensive variation in transcripts arising from alternative promoter usage, splicing, and polyadenylation. There are 16,247 new mouse protein-coding transcripts, including 5154 encoding previously unidentified proteins. Genomic mapping of the transcriptome reveals transcriptional forests, with overlapping transcription on both strands, separated by deserts in which few transcripts are observed. The data provide a comprehensive platform for the comparative analysis of mammalian transcriptional regulation in differentiation and development.

Analysis of the mouse transcriptome based on functional annotation of 60,770 full-length cDNAs

Only a small proportion of the mouse genome is transcribed into mature messenger RNA transcripts. There is an international collaborative effort to identify all full-length mRNA transcripts from the mouse, and to ensure that each is represented in a physical collection of clones. Here we report the manual annotation of 60,770 full-length mouse complementary DNA sequences. These are clustered into 33,409 'transcriptional units', contributing 90.1% of a newly established mouse transcriptome database. Of these transcriptional units, 4,258 are new protein-coding and 11,665 are new non-coding messages, indicating that non-coding RNA is a major component of the transcriptome. 41% of all transcriptional units showed evidence of alternative splicing. In protein-coding transcripts, 79% of splice variations altered the protein product. Whole-transcriptome analyses resulted in the identification of 2,431 sense-antisense pairs. The present work, completely supported by physical clones, provides the most comprehensive survey of a mammalian transcriptome so far, and is a valuable resource for functional genomics.

Regulation of actin dynamics through phosphorylation of cofilin by LIM-kinase

Cell division, cell motility and the formation and maintenance of specialized structures in differentiated cells depend directly on the regulated dynamics of the actin cytoskeleton. To understand the mechanisms of these basic cellular processes, the signalling pathways that link external signals to the regulation of the actin cytoskeleton need to be characterized. Here we identify a pathway for the regulation of cofilin, a ubiquitous actin-binding protein that is essential for effective depolymerization of actin filaments. LIM-kinase 1, also known as KIZ, is a protein kinase with two amino-terminal LIM motifs that induces stabilization of F-actin structures in transfected cells. Dominant-negative LIM-kinasel inhibits the accumulation of the F-actin. Phosphorylation experiments in vivo and in vitro provide evidence that cofilin is a physiological substrate of LIM-kinase 1. Phosphorylation by LIM-kinase 1 inactivates cofilin, leading to accumulation of actin filaments. Constitutively active Rac augmented cofilin phosphorylation and LIM-kinase 1 autophosphorylation whereas phorbol ester inhibited these processes. Our results define a mechanism for the regulation of cofilin and hence of actin dynamics in vivo. By modulating the stability of actin cytoskeletal structures, this pathway should play a central role in regulating cell motility and morphogenesis.

Crystal structure of an Hsp90-geldanamycin complex: targeting of a protein chaperone by an antitumor agent

The Hsp90 chaperone is required for the activation of several families of eukaryotic protein kinases and nuclear hormone receptors, many of which are protooncogenic and play a prominent role in cancer. The geldanamycin antibiotic has antiproliferative and antitumor effects, as it binds to Hsp90, inhibits the Hsp90-mediated conformational maturation/refolding reaction, and results in the degradation of Hsp90 substrates. The structure of the geldanamycin-binding domain of Hsp90 (residues 9-232) reveals a pronounced pocket, 15 A deep, that is highly conserved across species. Geldanamycin binds inside this pocket, adopting a compact structure similar to that of a polypeptide chain in a turn conformation. This, and the pocket's similarity to substrate-binding sites, suggest that the pocket binds a portion of the polypeptide substrate and participates in the conformational maturation/refolding reaction.

Genes specifically expressed at growth arrest of mammalian cells

A subtraction cDNA library enriched for RNA sequences preferentially expressed in growth-arrested cells was prepared. Six cDNA clones were identified, varying in abundance from 2% to 0.0002% of the library and in size from 0.8 to 10 kb. The corresponding mRNAs are downregulated with different kinetics upon induction of growth by serum. The kinetics of induction after serum starvation and density-dependent inhibition of two of these growth-arrest-specific (gas) genes were investigated in more detail. Two cell lines transformed by viral onc genes did not express the two gas genes. The full-length cDNA for one gene has been sequenced and the protein product preliminarily characterized by in vitro translation.

Overexpression of the neural growth-associated protein GAP-43 induces nerve sprouting in the adult nervous system of transgenic mice

Regulation of neurite outgrowth and structural plasticity may involve the expression of intrinsic determinants controlling growth competence. We have tested this concept by targeting constitutive expression of the growth-associated protein GAP-43 to the neurons of adult transgenic mice. Such mice showed striking spontaneous nerve sprouting at the neuromuscular junction and in the terminal field of hippocampal mossy fibers. In control mice, these nerve fibers did not express GAP-43, and did not sprout spontaneously. Lesion-induced nerve sprouting and terminal arborization during reinnervation were greatly potentiated in GAP-43-overexpressing mice. A mutant GAP-43 that cannot be phosphorylated by PKC had reduced sprout-promoting activity. The results establish GAP-43 as an intrinsic presynaptic determinant for neurite outgrowth and plasticity.

Ubiquitous cell-surface glycoprotein on tumor cells is proliferation-associated receptor for transferrin

A murine monoclonal antibody (OKT9) raised against human leukemic cells binds to a wide variety of leukemia and tumor cell lines and to a minority of leukemia cells taken directly from patients. Fetal thymus and liver are strongly reactive as are some normal, immature hemopoietic cells and activated lymphocytes. Reactivity with OKT9 appears to correlate with proliferation status in both normal and malignant populations. Biochemical analysis indicates that this structure is a approximately equal to 180,000-dalton glycoprotein with two disulfide-bonded subunits of approximately equal to 90,000-daltons. Isolation of the transferrin receptor from a T-cell line (MOLT-4) indicates that it also has a dimeric approximately equal to 180,000-dalton structure. Radio-labeled transferrin bound to its receptors can be specifically precipitated by the monoclonal OKT9, although the latter does not bind transferrin itself, indicating that the antigenic structure defined by this antibody is likely to be the transferrin receptor.

Convenient single-step, one tube purification of PCR products for direct sequencing

Images

Two distinct pathways of formation of 4-hydroxynonenal. Mechanisms of nonenzymatic transformation of the 9- and 13-hydroperoxides of linoleic acid to 4-hydroxyalkenals

The mechanism of formation of 4-hydroxy-2E-nonenal (4-HNE) has been a matter of debate since it was discovered as a major cytotoxic product of lipid peroxidation in 1980. Recent evidence points to 4-hydroperoxy-2E-nonenal (4-HPNE) as the immediate precursor of 4-HNE (Lee, S. H., and Blair, I. A. (2000) Chem. Res. Toxicol. 13, 698-702; Noordermeer, M. A., Feussner, I., Kolbe, A., Veldink, G. A., and Vliegenthart, J. F. G. (2000) Biochem. Biophys. Res. Commun. 277, 112-116), and a pathway via 9-hydroperoxylinoleic acid and 3Z-nonenal is recognized in plant extracts. Using the 9- and 13-hydroperoxides of linoleic acid as starting material, we find that two distinct mechanisms lead to the formation of 4-H(P)NE and the corresponding 4-hydro(pero)xyalkenal that retains the original carboxyl group (9-hydroperoxy-12-oxo-10E-dodecenoic acid). Chiral analysis revealed that 4-HPNE formed from 13S-hydroperoxy-9Z,11E-octadecadienoic acid (13S-HPODE) retains >90% S configuration, whereas it is nearly racemic from 9S-hydroperoxy-10E,12Z-octadecadienoic acid (9S-HPODE). 9-Hydroperoxy-12-oxo-10E-dodecenoic acid is >90% S when derived from 9S-HPODE and almost racemic from 13S-HPODE. Through analysis of intermediates and products, we provide evidence that (i) allylic hydrogen abstraction at C-8 of 13S-HPODE leads to a 10,13-dihydroperoxide that undergoes cleavage between C-9 and C-10 to give 4S-HPNE, whereas direct Hock cleavage of the 13S-HPODE gives 12-oxo-9Z-dodecenoic acid, which oxygenates to racemic 9-hydroperoxy-12-oxo-10E-dodecenoic acid; by contrast, (ii) 9S-HPODE cleaves directly to 3Z-nonenal as a precursor of racemic 4-HPNE, whereas allylic hydrogen abstraction at C-14 and oxygenation to a 9,12-dihydroperoxide leads to chiral 9S-hydroperoxy-12-oxo-10E-dodecenoic acid. Our results distinguish two major pathways to the formation of 4-HNE that should apply also to other fatty acid hydroperoxides. Slight ( approximately 10%) differences in the observed chiralities from those predicted in the above mechanisms suggest the existence of additional routes to the 4-hydroxyalkenals.

Structure of a Bag/Hsc70 complex: convergent functional evolution of Hsp70 nucleotide exchange factors

Bag (Bcl2-associated athanogene) domains occur in a class of cofactors of the eukaryotic chaperone 70-kilodalton heat shock protein (Hsp70) family. Binding of the Bag domain to the Hsp70 adenosine triphosphatase (ATPase) domain promotes adenosine 5'-triphosphate-dependent release of substrate from Hsp70 in vitro. In a 1.9 angstrom crystal structure of a complex with the ATPase of the 70-kilodalton heat shock cognate protein (Hsc70), the Bag domain forms a three-helix bundle, inducing a conformational switch in the ATPase that is incompatible with nucleotide binding. The same switch is observed in the bacterial Hsp70 homolog DnaK upon binding of the structurally unrelated nucleotide exchange factor GrpE. Thus, functional convergence has allowed proteins with different architectures to trigger a conserved conformational shift in Hsp70 that leads to nucleotide exchange.

Pharmacologic shifting of a balance between protein refolding and degradation mediated by Hsp90

The role of the abundant stress protein Hsp90 in protecting cells against stress-induced damage is not well understood. The recent discovery that a class of ansamycin antibiotics bind specifically to Hsp90 allowed us to address this problem from a new angle. We find that mammalian Hsp90, in cooperation with Hsp70, p60, and other factors, mediates the ATP-dependent refolding of heat-denatured proteins, such as firefly luciferase. Failure to refold results in proteolysis. The ansamycins inhibit refolding, both in vivo and in a cell extract, by preventing normal dissociation of Hsp90 from luciferase, causing its enhanced degradation. This mechanism also explains the ansamycin-induced proteolysis of several protooncogenic protein kinases, such as Raf-1, which interact with Hsp90. We propose that Hsp90 is part of a quality control system that facilitates protein refolding or degradation during recovery from stress. This function is used by a limited set of signal transduction molecules for their folding and regulation under nonstress conditions. The ansamycins shift the mode of Hsp90 from refolding to degradation, and this effect is probably amplified for specific Hsp90 substrates.

Regulation of p53 activity in nuclear bodies by a specific PML isoform

Covalent modification of the promyelocytic leukaemia protein (PML) by SUMO-1 is a prerequisite for the assembly of nuclear bodies (NBs), subnuclear structures disrupted in various human diseases and linked to transcriptional and growth control. Here we demonstrate that p53 is recruited into NBs by a specific PML isoform (PML3) or by coexpression of SUMO-1 and hUbc9. NB targeting depends on the direct association of p53, through its core domain, with a C-terminal region of PML3. The relocalization of p53 into NBs enhances p53 transactivation in a promoter-specific manner and affects cell survival. Our results indicate the existence of a cross-talk between PML- and p53-dependent growth suppression pathways, implying an important role for NBs and their resident proteins as modulators of p53 functions.

Primary structure of human transferrin receptor deduced from the mRNA sequence

In vertebrates all iron is taken up via the carrier protein transferrin. The carrier first binds its receptor and the receptor-ligand complex is then internalized via coated pits. The transferrin receptor is a transmembrane glycoprotein (apparent molecular weight (MW) 180,000) composed of two disulphide-bonded sub-units (each of apparent MW 90,000) It contains three N-linked glycan units and is post-translationally modified with both phosphate and fatty acyl groups. Here we have determined the nucleotide sequence of the coding region of the human transferrin receptor mRNA and from this deduced the amino acid sequence of the protein. The receptor does not contain an N-terminal signal peptide but there is a membrane-spanning segment 62 amino acids from the N-terminus. It therefore has a somewhat unusual configuration with a small N-terminal cytoplasmic domain and a C-terminal extracellular domain of 672 amino acids.

Myotonic dystrophy type 2: molecular, diagnostic and clinical spectrum

BACKGROUND

Myotonic dystrophy types 1 (DM1) and 2 (DM2/proximal myotonic myopathy PROMM) are dominantly inherited disorders with unusual multisystemic clinical features. The authors have characterized the clinical and molecular features of DM2/PROMM, which is caused by a CCTG repeat expansion in intron 1 of the zinc finger protein 9 (ZNF9) gene.

METHODS

Three-hundred and seventy-nine individuals from 133 DM2/PROMM families were evaluated genetically, and in 234 individuals clinical and molecular features were compared.

RESULTS

Among affected individuals 90% had electrical myotonia, 82% weakness, 61% cataracts, 23% diabetes, and 19% cardiac involvement. Because of the repeat tract's unprecedented size (mean approximately 5,000 CCTGs) and somatic instability, expansions were detectable by Southern analysis in only 80% of known carriers. The authors developed a repeat assay that increased the molecular detection rate to 99%. Only 30% of the positive samples had single sizeable expansions by Southern analysis, and 70% showed multiple bands or smears. Among the 101 individuals with single expansions, repeat size did not correlate with age at disease onset. Affected offspring had markedly shorter expansions than their affected parents, with a mean size difference of -17 kb (-4,250 CCTGs).

CONCLUSIONS

DM2 is present in a large number of families of northern European ancestry. Clinically, DM2 resembles adult-onset DM1, with myotonia, muscular dystrophy, cataracts, diabetes, testicular failure, hypogammaglobulinemia, and cardiac conduction defects. An important distinction is the lack of a congenital form of DM2. The clinical and molecular parallels between DM1 and DM2 indicate that the multisystemic features common to both diseases are caused by CUG or CCUG expansions expressed at the RNA level.

Studies on the corticospinal control of human walking. I. Responses to focal transcranial magnetic stimulation of the motor cortex

Experiments were done to determine the extent to which the corticospinal tract is linked with the segmental motor circuits controlling ankle flexors and extensors during human walking compared with voluntary motor tasks requiring attention to the level of motor activity. The motor cortex was activated transcranially using a focal magnetic stimulation coil. For each subject, the entire input-output (I-O) curve [i.e., the integral of the motor evoked-potential (MEP) versus stimulus strength] was measured during a prescribed tonic voluntary contraction of either the tibialis anterior (TA) or the soleus. Similarly, I-O curves were measured in the early part of the swing phase, or in the early part of the stance phase of walking. The I-O data points were fitted by the Boltzmann sigmoidal function, which accounted for >/=80% of total data variance. There was no statistically significant difference between the I-O curves of the TA measured during voluntary ankle dorsiflexion or during the swing phase of walking, at matched levels of background electromyographic (EMG) activity. Additionally, there was no significant difference in the relation between the coefficient of variation and the amplitude of the MEPs measured in each task, respectively. In comparison, during the stance phase of walking the soleus MEPs were reduced on average by 26% compared with their size during voluntary ankle plantarflexion. Furthermore, during stance the MEPs in the inactive TA were enhanced relative to their size during voluntary ankle plantarflexion and in four of six subjects the TA MEPs were larger than those of the soleus. Finally, stimulation of the motor cortex at various phases of the step cycle did not reset the cycle. The time of the next step occurred at the expected moment, as determined from the phase-resetting curve. One interpretation of this result is that the motor cortex may not be part of the central neural system involved in timing the motor bursts during the step cycle. We suggest that during walking the corticospinal tract is more closely linked with the segmental motor circuits controlling the flexor, TA, than it is with those controlling the extensor, soleus. However, during voluntary tasks requiring attention to the level of motor activity, it is equally linked with the segmental motor circuits of ankle flexors or extensors.

Reelin gene alleles and haplotypes as a factor predisposing to autistic disorder

Autistic disorder (MIM 209850) is currently viewed as a neurodevelopmental disease. Reelin plays a pivotal role in the development of laminar structures including the cerebral cortex, hippocampus, cerebellum and of several brainstem nuclei. Neuroanatomical evidence is consistent with Reelin involvement in autistic disorder. In this study, we describe several polymorphisms identified using RNA-SSCP and DNA sequencing. Association and linkage were assessed comparing 95 Italian patients to 186 ethnically-matched controls, and using the transmission/disequilibrium test and haplotype-based haplotype relative risk in 172 complete trios from 165 families collected in Italy and in the USA. Both case-control and family-based analyses yield a significant association between autistic disorder and a polymorphic GGC repeat located immediately 5' of the reelin gene (RELN) ATG initiator codon, as well as with specific haplotypes formed by this polymorphism with two single-base substitutions located in a splice junction in exon 6 and within exon 50. Triplet repeats located in 5' untranslated regions (5'UTRs) are indicative of strong transcriptional regulation. Our findings suggest that longer triplet repeats in the 5'UTR of the RELN gene confer vulnerability to autistic disorder.

Sequence of mdr3 cDNA encoding a human P-glycoprotein

We have determined the sequence of the human mdr3 gene using cDNA derived from liver RNA. The mdr3 gene codes for a member of a family of membrane proteins, the P-glycoproteins, overproduced in many multi-drug-resistant (MDR) cell lines. Like its relatives, the protein encoded by mdr3 has a deduced Mr of 140,000, which is presumably increased by glycosylation after synthesis. The sequence consists of two similar halves, each with a series of six hydrophobic segments that may form a membrane channel. The halves also possess nucleotide-binding consensus sequences, which presumably act as ATPases and drive drug transport. The presumed ATPase domains are all but identical to those of the human mdr1 gene product [Chen et al., Cell 47 (1986) 381-389]. We attribute this high level of sequence conservation to the repeated gene conversion that is evident from segments in which mdr1 and mdr3 differ only in a few silent mutations. Divergence between P-glycoprotein family members is greatest at the N terminus and in the 60 amino acid linker connecting the two halves. In the putative trans-membrane domains approx. 80% of the amino acids are conserved between the products of mdr1 and mdr3. Although the function of mdr3 is not yet known, its high homology with mdr1 suggests that it also encodes an efflux pump with broad specificity.

Development of Z-scores for fetal cardiac dimensions from echocardiography

OBJECTIVES

Z-scores for cardiac dimensions are well established in postnatal life, but have yet to be developed for fetal cardiac dimensions. These would be of real advantage to the clinician in accurately quantifying size and growth of cardiac dimensions and to the researcher by allowing mathematical comparison of growth in differing subgroups of a disease. The purpose of this observational study, conducted at tertiary fetal medicine and cardiology units, was to produce formulae and nomograms allowing computation of Z-scores for fetal cardiac dimensions from knowledge of femur length (FL), biparietal diameter (BPD) or gestational age (GA) using fetal echocardiography.

METHODS

Seventeen fetal cardiac dimensions were measured in 130 pregnant women with singleton fetuses of gestational age 15-39 weeks. Regression equations were derived relating all dimensions to FL, BPD and GA. From the calculations, formulae were then developed allowing fetal cardiac Z-score computation.

RESULTS

The relationships between cardiac dimensions and FL, BPD or GA were described following natural log transformation. From this analysis, FL (taken as an expression of fetal size) had the highest correlation to fetal cardiac dimensions. From the developed nomograms, Z-scores of specific fetal cardiac structures could be estimated from knowledge of the FL, BPD or GA and echocardiographically derived measurements.

CONCLUSIONS

This study allowed computation of Z-scores in fetal life for 17 cardiac dimensions from FL, BPD or GA. Previous studies of normal data allowed qualitative assessment of where abnormal cardiac dimensions lay with regard to the normal range. Z-scores from this study allow quantitative analysis of where such dimensions lie relative to the mean. This permits exact assessment of growth of fetal cardiac structures in normal hearts and particularly in congenitally abnormal hearts where quantitative assessment of the growth of cardiac structures is important in analyzing and planning treatment strategies.

Monoclonal antibodies OKT 11 and OKT 11A have pan-T reactivity and block sheep erythrocyte "receptors"

Monoclonal antibodies OKT11 (gamma 1) and OKT11A (gamma 2) are described and appear to have similar binding specificities. They bind, in immunofluorescence, with greater than 95% of infant thymocytes, staining both cortical and medullary cells, 65-80% of blood lymphocytes and selectively stain the T cell-dependent paracortical areas of tonsil. A small proportion (9-12%) of bone marrow lymphocytes stain, but this population excludes the terminal transferase-positive cells. Both the gamma 1 and gamma 2 antibodies stain the surface membrane Ig-negative lymphocytes in blood and tonsil and are to block sheep E rosette formation (to normal or leukemic T cells). In contrast, other monoclonal anti-T reagents tested (OKT1, OKT3, OKT4, OKT6, OKT8, OKT9, OKT10) did not block E rosette formation. E rosette formation and OKT11 bindings are coincident on T-ALL cell lines and both are trypsin-sensitive. In a series of 145 leukemias and 26 leukemic cell lines investigated, only leukemias with a T cell phenotype including E rosette positivity were reactive with OKT11 and OKT11A. OKT11A binds to a polypeptide of approximately 50 000 molecular weight on thymic lymphocytes. This structure may carry the recognition site for sheep erythrocytes. These antibodies provide additional useful markers for T cell analysis and are of potential therapeutic value.

Activated Ras interacts with the Ral guanine nucleotide dissociation stimulator

The yeast two-hybrid system was used to identify proteins that interact with Ras. The H-Ras protein was found to interact with a guanine nucleotide dissociation stimulator (GDS) that has been previously shown to regulate guanine nucleotide exchange on another member of the Ras protein family, Ral. The interaction is mediated by the C-terminal, noncatalytic segment of the RalGDS and can be detected both in vivo, using the two-hybrid system, and in vitro, with purified recombinant proteins. The interaction of the RalGDS C-terminal segment with Ras is specific, dependent on activation of Ras by GTP, and blocked by a mutation that affects Ras effector function. These characteristics are similar to those previously demonstrated for the interaction between Ras and its putative effector, Raf, suggesting that the RalGDS may also be a Ras effector. Consistent with this idea, the RalGDS was found to inhibit the binding of Raf to Ras.

The growth arrest-specific gene, gas1, is involved in growth suppression

This report describes the structure of the mRNA, the protein product, and the growth-regulating activity of one of the growth arrest-specific genes, gas1. From the predicted amino acid sequence, in vitro translation of gas1 mRNA, and immunofluorescence of cells in culture, it appears that the gas1 protein is an integral plasma membrane protein whose expression is linked to growth arrest. When gas1 is overexpressed from a constitutive promoter in quiescent cells, the serum-induced transition from the G0 to the S phase of the cell cycle is inhibited without affecting the normal early serum response. Ectopic expression of the gas1 gene by microinjection in normal and transformed NIH 3T3 cell lines with the notable exception of SV40-transformed 3T3 cells leads to inhibition of DNA synthesis. Thus, gas1 appears to be one component of a negative circuit that governs growth suppression. Its effect is, however, abolished in SV40-transformed cells.

Bone morphogenetic proteins are required in vivo for the generation of sympathetic neurons

Bone morphogenetic proteins (BMPs) induce autonomic neurogenesis in neural crest cultures and stimulate sympathetic neuron development when overexpressed in vivo. We demonstrate that inhibition of BMPs in the chick embryo bythe BMP antagonist Noggin prevents sympathetic neuron generation. In Noggin-treated embryos, the noradrenergic marker genes tyrosine hydroxylase (TH) and dopamine-beta-hydroxylase (DBH), panneuronal neurofilament 160 (NF160) and SCG10 genes, and the transcriptional regulators Phox2b and Phox2a are not expressed in sympathetic ganglia. Whereas initial ganglion development is not affected, the expression of the basic helix-loop-helix transcription factor Cash-1 is strongly reduced. These results demonstrate that BMPs are essential for sympathetic neuron development and establish Cash-1 and Phox2 genes as downstream effectors of BMPs in this lineage.

High-efficiency full-length cDNA cloning by biotinylated CAP trapper

We have devised a method for efficiently constructing high-content full-length cDNA libraries based on chemical introduction of a biotin group into the diol residue of the cap structure of eukaryotic mRNA, followed by RNase I treatment to select full-length cDNA. The selection occurs by trapping the biotin residue at the cap sites using streptavidin-coated magnetic beads, thus eliminating incompletely synthesized cDNAs. When this method was used to construct a mouse brain full-length cDNA library, our evaluation showed that more than 95% of the total clones were of full length, and recombinant clones could be produced with high efficiency (1.2 x 10(7)/10 micrograms starting mRNA). The analysis of 120 randomly picked clones indicates an unbiased representation of the starting mRNA population.