| 1. |
- Romagnoni, A, et al.
(author)
-
Comparative performances of machine learning methods for classifying Crohn Disease patients using genome-wide genotyping data
- 2019
-
In: Scientific reports. - : Springer Science and Business Media LLC. - 2045-2322. ; 9:1, s. 10351-
-
Journal article (peer-reviewed)abstract
- Crohn Disease (CD) is a complex genetic disorder for which more than 140 genes have been identified using genome wide association studies (GWAS). However, the genetic architecture of the trait remains largely unknown. The recent development of machine learning (ML) approaches incited us to apply them to classify healthy and diseased people according to their genomic information. The Immunochip dataset containing 18,227 CD patients and 34,050 healthy controls enrolled and genotyped by the international Inflammatory Bowel Disease genetic consortium (IIBDGC) has been re-analyzed using a set of ML methods: penalized logistic regression (LR), gradient boosted trees (GBT) and artificial neural networks (NN). The main score used to compare the methods was the Area Under the ROC Curve (AUC) statistics. The impact of quality control (QC), imputing and coding methods on LR results showed that QC methods and imputation of missing genotypes may artificially increase the scores. At the opposite, neither the patient/control ratio nor marker preselection or coding strategies significantly affected the results. LR methods, including Lasso, Ridge and ElasticNet provided similar results with a maximum AUC of 0.80. GBT methods like XGBoost, LightGBM and CatBoost, together with dense NN with one or more hidden layers, provided similar AUC values, suggesting limited epistatic effects in the genetic architecture of the trait. ML methods detected near all the genetic variants previously identified by GWAS among the best predictors plus additional predictors with lower effects. The robustness and complementarity of the different methods are also studied. Compared to LR, non-linear models such as GBT or NN may provide robust complementary approaches to identify and classify genetic markers.
|
|
| 2. |
- Feng, Shaohong, et al.
(author)
-
Dense sampling of bird diversity increases power of comparative genomics
- 2020
-
In: Nature. - : Springer Science and Business Media LLC. - 0028-0836 .- 1476-4687. ; 587:7833
-
Journal article (peer-reviewed)abstract
- Whole-genome sequencing projects are increasingly populating the tree of life and characterizing biodiversity(1-4). Sparse taxon sampling has previously been proposed to confound phylogenetic inference(5), and captures only a fraction of the genomic diversity. Here we report a substantial step towards the dense representation of avian phylogenetic and molecular diversity, by analysing 363 genomes from 92.4% of bird families-including 267 newly sequenced genomes produced for phase II of the Bird 10,000 Genomes (B10K) Project. We use this comparative genome dataset in combination with a pipeline that leverages a reference-free whole-genome alignment to identify orthologous regions in greater numbers than has previously been possible and to recognize genomic novelties in particular bird lineages. The densely sampled alignment provides a single-base-pair map of selection, has more than doubled the fraction of bases that are confidently predicted to be under conservation and reveals extensive patterns of weak selection in predominantly non-coding DNA. Our results demonstrate that increasing the diversity of genomes used in comparative studies can reveal more shared and lineage-specific variation, and improve the investigation of genomic characteristics. We anticipate that this genomic resource will offer new perspectives on evolutionary processes in cross-species comparative analyses and assist in efforts to conserve species. A dataset of the genomes of 363 species from the Bird 10,000 Genomes Project shows increased power to detect shared and lineage-specific variation, demonstrating the importance of phylogenetically diverse taxon sampling in whole-genome sequencing.
|
|
| 3. |
- Martin, Alec, et al.
(author)
-
UV-bright Star-forming Clumps and Their Host Galaxies in UVCANDELS at 0.5 ≤ z ≤ 1
- 2023
-
In: Astrophysical Journal. - 0004-637X .- 1538-4357. ; 955:2
-
Journal article (peer-reviewed)abstract
- Giant star-forming clumps are a prominent feature of star-forming galaxies (SFGs) and contain important clues on galaxy formation and evolution. However, the basic demographics of clumps and their host galaxies remain uncertain. Using the Hubble Space Telescope/Wide Field Camera 3 F275W images from the Ultraviolet Imaging of the Cosmic Assembly Near-infrared Deep Extragalactic Legacy Survey, we detect and analyze giant star-forming clumps in galaxies at 0.5 ≤ z ≤ 1, connecting two epochs when clumps are common (at cosmic high noon, z ∼ 2) and rare (in the local Universe). We construct a clump sample whose rest-frame 1600 Å luminosity is 3 times higher than the most luminous local H ii regions (MUV ≤ −16 AB). In our sample, 35% ± 3% of low-mass galaxies (log[M∗/M⊙] < 10) are clumpy (i.e., containing at least one off-center clump). This fraction changes to 22% ± 3% and 22% ± 4% for intermediate (10 ≤ log[M∗/M⊙] ≤ 10.5) and high-mass (log[M∗/M⊙] > 10.5) galaxies, in agreement with previous studies. When compared to similar-mass nonclumpy SFGs, low- and intermediate-mass clumpy SFGs tend to have higher star formation rates (SFRs) and bluer rest-frame U − V colors, while high-mass clumpy SFGs tend to be larger than nonclumpy SFGs. However, clumpy and nonclumpy SFGs have similar Sérsic index, indicating a similar underlying density profile. Furthermore, we investigate how the UV luminosity of star-forming regions correlates with the physical properties of host galaxies. On average, more luminous star-forming regions reside in more luminous, smaller, and/or higher specific SFR galaxies and are found closer to their hosts' galactic centers.
|
|
| 4. |
- Mehta, Vihang, et al.
(author)
-
A Spatially Resolved Analysis of Star Formation Burstiness by Comparing UV and Hα in Galaxies at z ∼ 1 with UVCANDELS
- 2023
-
In: Astrophysical Journal. - 0004-637X .- 1538-4357. ; 952:2
-
Journal article (peer-reviewed)abstract
- The UltraViolet imaging of the Cosmic Assembly Near-infrared Deep Extragalactic Legacy Survey Fields (UVCANDELS) program provides Hubble Space Telescope (HST)/UVIS F275W imaging for four CANDELS fields. We combine this UV imaging with existing HST/near-IR grism spectroscopy from 3D-HST+AGHAST to directly compare the resolved rest-frame UV and Hα emission for a sample of 979 galaxies at 0.7 < z < 1.5, spanning a range in stellar mass of 108−11.5M⊙. Using a stacking analysis, we perform a resolved comparison between homogenized maps of rest-UV and Hα to compute the average UV-to-Hα luminosity ratio (an indicator of burstiness in star formation) as a function of galactocentric radius. We find that galaxies below stellar mass of ∼109.5M⊙, at all radii, have a UV-to-Hα ratio higher than the equilibrium value expected from constant star formation, indicating a significant contribution from bursty star formation. Even for galaxies with stellar mass ≳109.5M⊙, the UV-to-Hα ratio is elevated toward their outskirts (R/Reff > 1.5), suggesting that bursty star formation is likely prevalent in the outskirts of even the most massive galaxies, but is likely overshadowed by their brighter cores. Furthermore, we present the UV-to-Hα ratio as a function of galaxy surface brightness, a proxy for stellar mass surface density, and find that regions below ∼107.5M⊙ kpc−2 are consistent with bursty star formation, regardless of their galaxy stellar mass, potentially suggesting that local star formation is independent of global galaxy properties at the smallest scales. Last, we find galaxies at z > 1.1 to have bursty star formation, regardless of radius or surface brightness.
|
|
| 5. |
- Sattari, Zahra, et al.
(author)
-
Fraction of Clumpy Star-forming Galaxies at 0.5 ≤ z ≤ 3 in UVCANDELS : Dependence on Stellar Mass and Environment
- 2023
-
In: Astrophysical Journal. - 0004-637X .- 1538-4357. ; 951:2
-
Journal article (peer-reviewed)abstract
- High-resolution imaging of galaxies in rest-frame UV has revealed the existence of giant star-forming clumps prevalent in high-redshift galaxies. Studying these substructures provides important information about their formation and evolution and informs theoretical galaxy evolution models. We present a new method to identify clumps in galaxies' high-resolution rest-frame UV images. Using imaging data from CANDELS and UVCANDELS, we identify star-forming clumps in an HST/F160W ≤ 25 AB mag sample of 6767 galaxies at 0.5 ≤ z ≤ 3 in four fields, GOODS-N, GOODS-S, EGS, and COSMOS. We use a low-passband filter in Fourier space to reconstruct the background image of a galaxy and detect small-scale features (clumps) on the background-subtracted image. Clumpy galaxies are defined as those having at least one off-center clump that contributes a minimum of 10% of the galaxy's total rest-frame UV flux. We measure the fraction of clumpy galaxies (fclumpy) as a function of stellar mass, redshift, and galaxy environment. Our results indicate that fclumpy increases with redshift, reaching ∼65% at z ∼ 1.5. We also find that fclumpy in low-mass galaxies () is 10% higher compared to that of their high-mass counterparts (). Moreover, we find no evidence of significant environmental dependence of fclumpy for galaxies at the redshift range of this study. Our results suggest that the fragmentation of gas clouds under violent disk instability remains the primary driving mechanism for clump formation, and incidents common in dense environments, such as mergers, are not the dominant processes.
|
|
