dnm_cohorts

This small package collects sample information and de novo mutations from published studies of de novo mutations in rare disease.

This also reconciles individuals and variants included in multiple studies, so we don't double count individuals if they are present in multiple studies.

All of the information is obtained from the relevant supplementary data tables associated with each journal publication. Where a study has collected healthy controls (e.g. unaffected siblings), the phenotype of the healthy controls is 'unaffected'.

Install

pip install git+git://github.com/jeremymcrae/dnm_cohorts.git

Usage

The data files are included in the package, and can be loaded in python with:

from dnm_cohorts import de_novos, cohort

Build data files

# to create a table of all individuals in the cohorts
dnm_cohorts --cohorts --output test.txt

# to create a table of all de novo mutations found in those individuals
dnm_cohorts --de-novos --output test.txt

The package contains a dataset of de novos on their original genome build (the default), as well as a dataset lifted to GRCh37 and a dataset lifted to GRCh38.

Cohorts

reference	year	unique individuals	phenotype	assay	deprecated
De Ligt et al. N Engl J Med 367:1921-1929	2012	100	intellectual disability	exome	no
Iossifov et al. Neuron 74:285-299	2012	686 (343 asd, 343 unaffected)	autism spectrum disorder, unaffected siblings	exome	no
O'Roak et al. Nature 485:246-250	2012	206	autism spectrum disorder	exome	no
Rauch et al. Lancet 380:1674-1682	2012	51	intellectual disability	exome	no
Sanders et al. Nature 485:237-241	2012	452 (238 asd, 214 unaffected)	autism spectrum disorder, unaffected siblings	exome	no
Epi4K Consortium. Nature 501:217-221	2013	264	epilepsy	exome	no
EuroEPINOMICS-RES Consortium. AJHG 95:360-370	2014	92	epilepsy	exome	no
Gilissen et al. Nature 511:344-347	2014	0	intellectual disability	exome	no, but only extends De ligt et al
De Rubeis et al. Nature 515:209-215	2014	1604 (1443 asd, 161 unaffected)	autism spectrum disorder, unaffected siblings	exome	no
Iossifov et al. Nature 498:216-221	2014	3095 (1726 asd, 1369 unaffected)	autism spectrum disorder, unaffected siblings	exome	no
Sanders et al. Neuron 87:1215-1233	2015	750 (314 asd, 436 unaffected)	autism spectrum disorder, unaffected siblings	exome	no
Homsy et al. Science 350:1262-1266	2015	1213	congenital heart disease	exome	yes, superseded by Jin et al cohort
Lelieveld et al. Nature Neuroscience 19:1194-1196	2016	820	intellectual disability	exome	yes, superseded by Kaplanis et al cohort
McRae et al. Nature 542:433-438	2017	4293	intellectual disability	exome	yes, superseded by Kaplanis et al cohort
Jónsson et al. Nature 549:519-522	2017	1458	unaffected	genome (no chrX in males)	yes
Jin et al. Nature Genetics 49:1593-1601	2017	2465	congenital heart disease	exome	no
Yuen et al. Nature Neuroscience 20:602-611	2017	5265	autism spectrum disorder	genome	no
An et al. Science 362:eaat6576	2018	1902 affected, 1902 unaffected	autism spectrum disorder, unaffected siblings	genome	no
Halldorsson et al. Science	2019	2976	unaffected, supersedes Jónsson et al	genome	no
Kaplanis et al. Nature	2019	31058	intellectual disability, supersedes Lelieveld et al and McRae et al cohorts	exome	no
Fu et al. Nature Genetics 54:1320-1331	2022	42607	autism spectrum disorder	exome (no chrX)	no, but Fu et al and Zhou et al mostly overlap
Zhou et al. Nature Genetics 54:1305-1319	2022	42607	autism spectrum disorder	exome	yes, by Fu et al, because Fu et al include all sample IDs

Excluded cohorts

We exclude some published cohorts with de novo mutations, for the reasons below:

• Helbig et al in Genetics in Medicine 18:898-905. This study only reported the likely pathogenic de novo mutations.
• Goldman et al. in Nature Genetics 48:935-939. This study includes monozygotic twins, so some de novo mutations are not independent, but does not include sample or family IDs that would permit exclusion of the monozygotic twins.
• Goldmann et al. in Nature Genetics 50:487-492. This study only reported clustered DNMs, so not representative of all coding DNMs.