| 1. |
|
|
| 2. |
|
|
| 3. |
- Bäcklin, Christofer L., 1983-, et al.
(författare)
-
Developer-Friendly and Computationally Efficient Predictive Modeling without Information Leakage : The emil Package for R
- 2018
-
Ingår i: Journal of Statistical Software. - : JOURNAL STATISTICAL SOFTWARE. - 1548-7660. ; 85:13, s. 1-30
-
Tidskriftsartikel (refereegranskat)abstract
- Data driven machine learning for predictive modeling problems (classification, regression, or survival analysis) typically involves a number of steps beginning with data preprocessing and ending with performance evaluation. A large number of packages providing tools for the individual steps are available for R, but there is a lack of tools for facilitating rigorous performance evaluation of the complete procedures assembled from them by means of cross-validation, bootstrap, or similar methods. Such a tool should strictly prevent test set observations from influencing model training and meta- parameter tuning, so- called information leakage, in order to not produce overly optimistic performance estimates. Here we present a new package for R denoted emil (evaluation of modeling without information leakage) that offers this form of performance evaluation. It provides a transparent and highly customizable framework for facilitating the assembly, execution, performance evaluation, and interpretation of complete procedures for classification, regression, and survival analysis. The components of package emil have been designed to be as modular and general as possible to allow users to combine, replace, and extend them if needed. Package emil was also developed with scalability in mind and has a small computational overhead, which is a key requirement for analyzing the very big data sets now available in fields like medicine, physics, and finance. First package emil's functionality and usage is explained. Then three specific application examples are presented to show its potential in terms of parallelization, customization for survival analysis, and development of ensemble models. Finally a brief comparison to similar software is provided.
|
|
| 4. |
- Bäcklin, Christofer, 1983-
(författare)
-
Machine Learning Based Analysis of DNA Methylation Patterns in Pediatric Acute Leukemia
- 2015
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Acute lymphoblastic leukemia (ALL) is the most common pediatric cancer in the Nordic countries. Recent evidence indicate that DNA methylation (DNAm) play a central role in the development and progression of the disease.DNAm profiles of a collection of ALL patient samples and a panel of non-leukemic reference samples were analyzed using the Infinium 450k methylation assay. State-of-the-art machine learning algorithms were used to search the large amounts of data produced for patterns predictive of future relapses, in vitro drug resistance, and cytogenetic subtypes, aiming at improving our understanding of the disease and ultimately improving treatment.In paper I, the predictive modeling framework developed to perform the analyses of DNAm dataset was presented. It focused on uncompromising statistical rigor and computational efficiency, while allowing a high level of modeling flexibility and usability. In paper II, the DNAm landscape of ALL was comprehensively characterized, discovering widespread aberrant methylation at diagnosis strongly influenced by cytogenetic subtype. The aberrantly methylated regions were enriched for genes repressed by polycomb group proteins, repressively marked histones in healthy cells, and genes associated with embryonic development. A consistent trend of hypermethylation at relapse was also discovered. In paper III, a tool for DNAm-based subtyping was presented, validated using blinded samples and used to re-classify samples with incomplete phenotypic information. Using RNA-sequencing, previously undetected non-canonical aberrations were found in many re-classified samples. In paper IV, the relationship between DNAm and in vitro drug resistance was investigated and predictive signatures were obtained for seven of the eight therapeutic drugs studied. Interpretation was challenging due to poor correlation between DNAm and gene expression, further complicated by the discovery that random subsets of the array can yield comparable classification accuracy. Paper V presents a novel Bayesian method for multivariate density estimation with variable bandwidths. Simulations showed comparable performance to the current state-of-the-art methods and an advantage on skewed distributions.In conclusion, the studies characterize the information contained in the aberrant DNAm patterns of ALL and assess its predictive capabilities for future relapses, in vitro drug sensitivity and subtyping. They also present three publicly available tools for the scientific community to use.
|
|
| 5. |
- Bäcklin, Christofer, 1983-, et al.
(författare)
-
Self-tuning density estimation based on Bayesian averaging of adaptive kernel density estimations yields state-of-the-art performance
- 2018
-
Ingår i: Pattern Recognition. - : ELSEVIER SCI LTD. - 0031-3203 .- 1873-5142. ; 78, s. 133-143
-
Tidskriftsartikel (refereegranskat)abstract
- Non-parametric probability density function (pdf) estimation is a general problem encountered in many fields. A promising alternative to the dominating solutions, kernel density estimation (KDE) and Gaussian mixture modeling, is adaptive KDE where kernels are given individual bandwidths adjusted to the local data density. Traditionally the bandwidths are selected by a non-linear transformation of a pilot pdf estimate, containing parameters controlling the scaling, but identifying parameters values yielding competitive performance has turned out to be non-trivial. We present a new self-tuning (parameter free) pdf estimation method called adaptive density estimation by Bayesian averaging (ADEBA) that approximates pdf estimates in the form of weighted model averages across all possible parameter values, weighted by their Bayesian posterior calculated from the data. ADEBA is shown to be simple, robust, competitive in comparison to the current practice, and easily generalize to multivariate distributions. An implementation of the method for R is publicly available.
|
|
| 6. |
- Carlsson Almlöf, Jonas, et al.
(författare)
-
Novel risk genes for systemic lupus erythematosus predicted by random forest classification
- 2017
-
Ingår i: Scientific Reports. - : Springer Science and Business Media LLC. - 2045-2322. ; 7:1
-
Tidskriftsartikel (refereegranskat)abstract
- Genome-wide association studies have identified risk loci for SLE, but a large proportion of the genetic contribution to SLE still remains unexplained. To detect novel risk genes, and to predict an individual's SLE risk we designed a random forest classifier using SNP genotype data generated on the "Immunochip" from 1,160 patients with SLE and 2,711 controls. Using gene importance scores defined by the random forest classifier, we identified 15 potential novel risk genes for SLE. Of them 12 are associated with other autoimmune diseases than SLE, whereas three genes (ZNF804A, CDK1, and MANF) have not previously been associated with autoimmunity. Random forest classification also allowed prediction of patients at risk for lupus nephritis with an area under the curve of 0.94. By allele-specific gene expression analysis we detected cis-regulatory SNPs that affect the expression levels of six of the top 40 genes designed by the random forest analysis, indicating a regulatory role for the identified risk variants. The 40 top genes from the prediction were overrepresented for differential expression in B and T cells according to RNA-sequencing of samples from five healthy donors, with more frequent over-expression in B cells compared to T cells.
|
|
| 7. |
- Darmanis, Spyros, et al.
(författare)
-
ProteinSeq : high-performance proteomic analyses by proximity ligation and next generation sequencing
- 2011
-
Ingår i: PLOS ONE. - : Public Library of Science (PLoS). - 1932-6203. ; 6:9, s. e25583-
-
Tidskriftsartikel (refereegranskat)abstract
- Despite intense interest, methods that provide enhanced sensitivity and specificity in parallel measurements of candidate protein biomarkers in numerous samples have been lacking. We present herein a multiplex proximity ligation assay with readout via realtime PCR or DNA sequencing (ProteinSeq). We demonstrate improved sensitivity over conventional sandwich assays for simultaneous analysis of sets of 35 proteins in 5 μl of blood plasma. Importantly, we observe a minimal tendency to increased background with multiplexing, compared to a sandwich assay, suggesting that higher levels of multiplexing are possible. We used ProteinSeq to analyze proteins in plasma samples from cardiovascular disease (CVD) patient cohorts and matched controls. Three proteins, namely P-selectin, Cystatin-B and Kallikrein-6, were identified as putative diagnostic biomarkers for CVD. The latter two have not been previously reported in the literature and their potential roles must be validated in larger patient cohorts. We conclude that ProteinSeq is promising for screening large numbers of proteins and samples while the technology can provide a much-needed platform for validation of diagnostic markers in biobank samples and in clinical use.
|
|
| 8. |
- Edfeldt, Katarina, 1979-, et al.
(författare)
-
DcR3, TFF3 and Midkine are Novel Serum Biomarkers in Small Intestinal Neuroendocrine Tumors
- 2017
-
Ingår i: Neuroendocrinology. - : S. Karger AG. - 0028-3835 .- 1423-0194. ; 105:2, s. 170-181
-
Tidskriftsartikel (refereegranskat)abstract
- Small intestinal neuroendocrine tumors (SI-NETs) are amine- and peptide producing neoplasms. Most patients display metastases at the time of diagnosis, they have an unpredictable individual disease course and the tumors are often therapy resistant. Chromogranin A (CgA) and 5-hydroxyindoleacetic acid (5-HIAA) are the clinically most used biomarkers today, but there is a great need for novel diagnostic and prognostic biomarkers and new therapeutic targets. Sixty-nine biomarkers were screened in serum from 23 SI-NET patients and 23 healthy controls using multiplex PLA (proximity ligation assay). A refined method, PEA (proximity extension assay), was used to analyze 76 additional biomarkers. Statistical testing and multivariate classification were performed. Immunohistochemistry and ELISA assays were performed in an extended cohort. Using PLA, 19 biomarkers showed a significant difference in serum concentrations between patients and controls, and PEA revealed difference in concentrations in 13 proteins. Multivariate classification analysis revealed decoy receptor 3 (DcR3), trefoil factor 3 (TFF3) and Midkine to be good biomarkers for disease, which was confirmed by ELISA analysis. All three biomarkers were expressed in tumor tissue. DcR3 concentrations were elevated in patients with stage IV disease. High concentrations of DcR3 and TFF3 were correlated to poor survival. DcR3, TFF3 and Midkine exhibited elevated serum concentrations in SI-NET patients compared to healthy controls, and DcR3 and TFF3 were associated with poor survival. DcR3 seems to be a marker for liver metastases while TFF3 and Midkine may be new diagnostic biomarkers for SI-NETs.
|
|
| 9. |
|
|
| 10. |
|
|
| 11. |
- Jacobsson Svärd, Staffan, et al.
(författare)
-
Non-destructive experimental determination of the pin-power distribution in nuclear fuel
- 2003
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- A need for validation of modern core-analysis codes with respect to the calculated pin-power distribution has been recognized. A non-destructive experimental method for such validation has been developed, based on a tomographic technique. Each axial node of the fuel assembly is measured separately and the relative pin-by-pin content of the direct fission product Ba-140 is determined. Investigations performed so far indicate that 1-2% (1 σ) accuracy can be obtained.A measuring device has been constructed which, when fully equipped, is designed to measure a complete BWR assembly in 25 axial nodes within an eight-hour work shift. The applicability of the constructed device has been demonstrated in measurements at the Swedish BWR Forsmark 2 on irradiated fuel with a cooling time of 4-5 weeks. Data from the core-analysis code POLCA-7 have been compared to measured pin-by-pin contents of Ba-140. An agreement of 3.1% (1 σ) has been demonstrated.As compared to the conventional method, involving gamma scanning of individual fuel pins, this method does not require the fuel to be disassembled. Neither does the fuel channel have to be removed. The cost per measured fuel pin is in the order of 20 times lower than the conventional method.
|
|
| 12. |
|
|
| 13. |
- Jacobsson Svärd, Staffan, et al.
(författare)
-
Nondestructive Experimental Determination of the Pin-Power Distribution in Nuclear Fuel Assemblies
- 2005
-
Ingår i: Nuclear Technology. - 0029-5450. ; 151:1, s. 70-76
-
Tidskriftsartikel (refereegranskat)abstract
- A need for validation of modern production codes with respect to the calculated pin-power distribution has been recognized. A nondestructive experimental method for such validation has been developed based on a tomographic technique. The gamma-ray flux distribution is recorded in each axial node of the fuel assembly separately, whereby the relative rod-by-rod content of the fission product 140Ba is determined. Measurements indicate that 1 to 2% accuracy (1 sigma) is achievable.A device has been constructed for in-pool measurements at reactor sites. The applicability has been demonstrated in measurements at the Swedish boiling water reactor (BWR) Forsmark 2 on irradiated fuel with a cooling time of 4 to 5 weeks. Data from the production code POLCA-7 have been compared to measured rod-by-rod contents of 140Ba. An agreement of 3.1% (1 sigma) has been demonstrated.It is estimated that measurements can be performed on a complete BWR assembly in 25 axial nodes within an 8-h work shift. As compared to the conventional method, involving gamma scanning of individual fuel rods, this method does not require the fuel to be disassembled nor does the fuel channel have to be removed. The cost per measured fuel rod is estimated to be an order of magnitude lower than the conventional method.
|
|
| 14. |
- Jacobsson Svärd, Staffan, et al.
(författare)
-
Tomography for partial-defect verification : experiences from measurements using different devices
- 2006
-
Ingår i: ESARDA Bulletin. - 0392-3029. ; 33, s. 15-25
-
Tidskriftsartikel (refereegranskat)abstract
- Three devices of different types have been used in tomographic measurements for the purpose of partial-defect verification on the single-rod level. The devices range from a laboratory device used in measurements on a fuel model to an in-pool device used in measurements on irradiated fuel in a fuel-handling pool.The tomographic technique accounted for in this paper involves measurements of the gamma-ray flux distribution around a fuel assembly followed by computer-aided reconstruction of the internal source distribution. The results are rod-by-rod values of the relative concentrations of selected gamma-emitting isotopes. Also cross-sectional images are obtained.The tomographic technique presented here has proven to be robust and reliable. In laboratory experiments on a fuel model, reconstructions of relative rod-by-rod activities have been obtained with 1.5 % accuracy (1 σ). Using an in-pool device in measurements on fuel with a cooling time of about 4 weeks, data on fuel rods have been obtained in agreement with production-code calculations. Furthermore, tomographic images of good quality have been acquired.The applicability of the tomographic technique for partial-defect verification on the single-rod level has been investigated and demonstrated. The gamma-ray source concentration reconstructed in a position corresponding to a removed or replaced rod has been significantly lower than that of normal rods.Finally, requirements and properties of a device for tomographic measurements on nuclear fuel are discussed. It is argued that the use of a detector system with high energy resolution and high peak efficiency in connection to spectroscopic peak analysis is beneficial.
|
|
| 15. |
- Krali, Olga, et al.
(författare)
-
Dna methylation signatures predict cytogenetic subtype and outcome in pediatric acute myeloid leukemia (Aml)
- 2021
-
Ingår i: Genes. - : MDPI AG. - 2073-4425. ; 12:6
-
Tidskriftsartikel (refereegranskat)abstract
- Pediatric acute myeloid leukemia (AML) is a heterogeneous disease composed of clinically relevant subtypes defined by recurrent cytogenetic aberrations. The majority of the aberrations used in risk grouping for treatment decisions are extensively studied, but still a large proportion of pediatric AML patients remain cytogenetically undefined and would therefore benefit from additional molecular investigation. As aberrant epigenetic regulation has been widely observed during leukemogenesis, we hypothesized that DNA methylation signatures could be used to predict molecular subtypes and identify signatures with prognostic impact in AML. To study genome-wide DNA methylation, we analyzed 123 diagnostic and 19 relapse AML samples on Illumina 450k DNA methylation arrays. We designed and validated DNA methylation-based classifiers for AML cytogenetic subtype, resulting in an overall test accuracy of 91%. Furthermore, we identified methylation signatures associated with outcome in t(8;21)/RUNX1-RUNX1T1, normal karyotype, and MLL/KMT2A-rearranged subgroups (p < 0.01). Overall, these results further underscore the clinical value of DNA methylation analysis in AML. © 2021 by the authors. Licensee MDPI, Basel, Switzerland.
|
|
| 16. |
- Marzouka, Nour-al-dain, et al.
(författare)
-
CopyNumber450kCancer : baseline correction for accurate copy number calling from the 450k methylation array
- 2016
-
Ingår i: Bioinformatics. - : Oxford University Press (OUP). - 1367-4803 .- 1367-4811. ; 32:7, s. 1080-1082
-
Tidskriftsartikel (refereegranskat)abstract
- The Illumina Infinium HumanMethylation450 BeadChip (450k) is widely used for the evaluation of DNA methylation levels in large-scale datasets, particularly in cancer. The 450k design allows copy number variant (CNV) calling using existing bioinformatics tools. However, in cancer samples, numerous large-scale aberrations cause shifting in the probe intensities and thereby may result in erroneous CNV calling. Therefore, a baseline correction process is needed. We suggest the maximum peak of probe segment density to correct the shift in the intensities in cancer samples.
|
|
| 17. |
|
|
| 18. |
- Nordlund, Jessica, et al.
(författare)
-
DNA methylation-based subtype prediction for pediatric acute lymphoblastic leukemia
- 2015
-
Ingår i: Clinical Epigenetics. - : Springer Science and Business Media LLC. - 1868-7083 .- 1868-7075. ; 7
-
Tidskriftsartikel (refereegranskat)abstract
- Background: We present a method that utilizes DNA methylation profiling for prediction of the cytogenetic subtypes of acute lymphoblastic leukemia (ALL) cells from pediatric ALL patients. The primary aim of our study was to improve risk stratification of ALL patients into treatment groups using DNA methylation as a complement to current diagnostic methods. A secondary aim was to gain insight into the functional role of DNA methylation in ALL. Results: We used the methylation status of similar to 450,000 CpG sites in 546 well-characterized patients with T-ALL or seven recurrent B-cell precursor ALL subtypes to design and validate sensitive and accurate DNA methylation classifiers. After repeated cross-validation, a final classifier was derived that consisted of only 246 CpG sites. The mean sensitivity and specificity of the classifier across the known subtypes was 0.90 and 0.99, respectively. We then used DNA methylation classification to screen for subtype membership of 210 patients with undefined karyotype (normal or no result) or non-recurrent cytogenetic aberrations('other' subtype). Nearly half (n = 106) of the patients lacking cytogenetic subgrouping displayed highly similar methylation profiles as the patients in the known recurrent groups. We verified the subtype of 20% of the newly classified patients by examination of diagnostic karyotypes, array-based copy number analysis, and detection of fusion genes by quantitative polymerase chain reaction (PCR) and RNA-sequencing (RNA-seq). Using RNA-seq data from ALL patients where cytogenetic subtype and DNA methylation classification did not agree, we discovered several novel fusion genes involving ETV6, RUNX1, and PAX5. Conclusions: Our findings indicate that DNA methylation profiling contributes to the clarification of the heterogeneity in cytogenetically undefined ALL patient groups and could be implemented as a complementary method for diagnosis of ALL. The results of our study provide clues to the origin and development of leukemic transformation. The methylation status of the CpG sites constituting the classifiers also highlight relevant biological characteristics in otherwise unclassified ALL patients.
|
|
| 19. |
- Nordlund, Jessica, et al.
(författare)
-
Genome-wide signatures of differential DNA methylation in pediatric acute lymphoblastic leukemia
- 2013
-
Ingår i: Genome Biology. - : Springer Science and Business Media LLC. - 1465-6906 .- 1474-760X. ; 14:9, s. r105-
-
Tidskriftsartikel (refereegranskat)abstract
- BACKGROUND:Although aberrant DNA methylation has been observed previously in acute lymphoblastic leukemia (ALL), the patterns of differential methylation have not been comprehensively determined in all subtypes of ALL on a genome-wide scale. The relationship between DNA methylation, cytogenetic background, drug resistance and relapse in ALL is poorly understood.RESULTS:We surveyed the DNA methylation levels of 435,941 CpG sites in samples from 764 children at diagnosis of ALL and from 27 children at relapse. This survey uncovered four characteristic methylation signatures. First, compared with control blood cells, the methylomes of ALL cells shared 9,406 predominantly hypermethylated CpG sites, independent of cytogenetic background. Second, each cytogenetic subtype of ALL displayed a unique set of hyper- and hypomethylated CpG sites. The CpG sites that constituted these two signatures differed in their functional genomic enrichment to regions with marks of active or repressed chromatin. Third, we identified subtype-specific differential methylation in promoter and enhancer regions that were strongly correlated with gene expression. Fourth, a set of 6,612 CpG sites was predominantly hypermethylated in ALL cells at relapse, compared with matched samples at diagnosis. Analysis of relapse-free survival identified CpG sites with subtype-specific differential methylation that divided the patients into different risk groups, depending on their methylation status.CONCLUSIONS:Our results suggest an important biological role for DNA methylation in the differences between ALL subtypes and in their clinical outcome after treatment.
|
|
| 20. |
- Osifo, Otasowie, et al.
(författare)
-
Verification and determination of the decay heat in spent PWR fuel by means of gamma scanning
- 2008
-
Ingår i: Nuclear science and engineering. - 0029-5639 .- 1943-748X. ; 160:1, s. 129-143
-
Tidskriftsartikel (refereegranskat)abstract
- Decay heat is an important design parameter at the future Swedish spent nuclear fuel repository. It will be calculated for each fuel assembly using dedicated depletion codes, based on the operator-declared irradiation history. However, experimental verification of the calculated decay heat is also anticipated. Such verification may, be obtained by, gamma scanning using the established correlation between the decay heat and the emitted gamma-ray intensity from Cs-137. In this procedure, the correctness of the operator-declared fuel parameters can be verified. Recent achievements of the gamma-scanning technique include the development of a dedicated spectroscopic data-acquisition system and the use of an advanced calorimeter for calibration. Using this system, the operator-declared burnup and cooling time of 31 pressurized water reactor fuel assemblies was verified experimentally, to within 2.2% (1 sigma) and 1.9% (1 sigma), respectively. The measured decay heat agreed with calorimetric data within 2.3% (1 sigma). whereby the calculated decay, heat was verified within 2.3% (1 sigma). The measuring time per fuel assembly was similar to 15 min. In case reliable operator-declared data are not available, the gamma-scanning technique also provides a means to independently measure the decay, heat. The results obtained in this procedure agreed with calorimetric data within 2.7% (1 sigma).
|
|
| 21. |
- Willman, Christofer, et al.
(författare)
-
A nondestructive method for discriminating MOX fuel from LEU fuel for safeguards purposes
- 2006
-
Ingår i: Annals of Nuclear Energy. - : Elsevier BV. - 0306-4549 .- 1873-2100. ; 33:9, s. 766-773
-
Tidskriftsartikel (refereegranskat)abstract
- Plutonium-rich mixed oxide fuel (MOX) is increasingly used in thermal reactors. However, spent MOX fuel could be a potential source of nuclear weapons material and a safeguards issue is therefore to determine whether a spent nuclear fuel assembly is of MOX type or of LEU (Low Enriched Uranium) type. In this paper, we present theoretical and experimental results of a study that aims to investigate the possibilities of using gamma-ray spectroscopy to determine whether a nuclear fuel assembly is of MOX or of LEU type. Simulations with the computer code ORIGEN-ARP have been performed where LEU and MOX fuel types with varying enrichment and burnup as well as different irradiation histories have been modelled. The simulations indicate that the fuel type determination may be achieved by using the intensity ratio Cs-134/Eu-154. An experimental study of MOX fuel of 14 x 14 PWR type and LEU fuel of both 15 x 15 and 17 x 17 type is also reported in this paper. The outcome of the experimental study support the conclusion that MOX fuel may be discriminated from LEU fuel by measuring the suggested isotopic ratio.
|
|
| 22. |
|
|
| 23. |
|
|
| 24. |
- Willman, Christofer, et al.
(författare)
-
Nondestructive assay of spent nuclear fuel with gamma-ray spectroscopy
- 2006
-
Ingår i: Annals of Nuclear Energy. - 0306-4549. ; 33:5, s. 427-438
-
Tidskriftsartikel (refereegranskat)abstract
- An important issue in nuclear safeguards is to verify operator-declared data of spent nuclear fuel. Various techniques have therefore been assigned for this purpose. A nondestructive approach is to measure the gamma radiation from spent nuclear fuel assemblies. Using this technique, parameters such as burnup and cooling time can be calculated or verified. In this paper, we propose the utilization of gamma rays from 137Cs, 134Cs and 154Eu to determine the consistency of operator-declared information. Specifically, we have investigated to what extent irradiation histories can be verified. Computer simulations were used in order to determine limits for detecting small deviations from declared data. In addition, the technique has been experimentally demonstrated on 12 PWR fuel assemblies. A technique for determining burnup and cooling time for fuel assemblies where no operator-declared information is available is also presented. In such a case, the burnup could be determined with 1.6% relative standard deviation and the cooling time with 1.5%.
|
|
