The week of February 21-25, 2022, we hosted
the primary Widespread Fund Information Ecosystem (CFDE) Hackathon. The
targets of the hackathon have been to extend familiarity with knowledge units
from Widespread Fund packages and
work in the direction of cross-cutting, integrative analyses.
We invited members of the CFDE to suggest hackathon classes to
introduce their Widespread Fund knowledge units and supply technical help
whereas attendees explored the info. Periods featured knowledge from the
CFDE Portal,
Human Microbiome Undertaking (HMP),
Extracellular RNA Communication (exRNA),
Metabolomics Workbench (MW),
and
Signature Commons Library of Built-in Community-Primarily based Mobile Signatures (SigCom LINCS).
The Hackathon Periods
All classes have been recorded and will be seen on the Session Particulars and Recordings web page of the hackathon web site!
This digital occasion started Monday morning with a welcome deal with by Dr. Titus Brown (UC Davis) adopted by displays from every Widespread Fund Program to offer a short overview of their knowledge and session targets.
On Monday afternoon, Dr. Amanda Charbonneau (UC Davis) taught
attendees methods to use the CFDE Portal
to search out datasets from taking part Widespread Fund
packages. Dr. Charbonneau used HMP knowledge as a motivating instance,
then helped attendees uncover knowledge units from different packages. These
datasets are fairly massive, so on Tuesday afternoon, Dr. Charbonneau
taught a second session on methods to obtain and course of knowledge from the
CFDE Portal utilizing Amazon Internet Companies (AWS). Attendees have been offered
with AWS accounts that they may use to investigate knowledge found
by means of the portal.
On Tuesday morning, Emily LaPlante and Keyang Yu (Baylor Faculty of
Medication) offered an outline of the
exRNA Atlas, which comprises over 7,500
small RNA sequences and qPCR profiles from human and mouse, and
launched attendees to quite a lot of software program instruments for exploring RNA
binding proteins. This session explored two use instances:
1) Discovering RNA binding proteins and their related RNA cargo in quite a lot of human biofluids and exploring their utility as biomarkers
2) Exploring different websites throughout the genome by intersecting exRNA Atlas
knowledge with areas of curiosity utilizing BedGraph recordsdata, in addition to
making use of this method to different datasets.
On Wednesday morning, Eoin Fahy and Mano Maurya (UCSD) launched the
Metabolomics Workbench
database which comprises over 164,000 molecular constructions overlaying
100+ species! Attendees realized methods to work together with the Metabolomics
Workbench Portal after which seen an illustration of
MetENP,
an R bundle that permits detection of serious metabolites from
metabolite data.
The ultimate data-driven hackathon session came about on Thursday afternoon. John
Erol Evangelista (Mt Sinai) launched the SigCom LINCS
API which comprises over 1.5 million gene expression signatures from LINCS,
the Gene Expression Tissue Undertaking (GTEx), and Gene Expression Omnibus
databases (GEO). Then, Daniel Clarke (Mt Sinai) gave an introduction to
constructing Appyters and methods to
use the SigCom LINCS APIs inside Appyters.
On Friday we ran a Wrap Up and Future Instructions session for
presenters to recap what occurred at their classes and discuss
future targets for his or her instruments. This allowed everybody to find out about
classes they may not have attended, and presumably spark curiosity in
watching the video recording of the session.
Reflection
General the classes have been nicely attended and nicely obtained! In a
pre-hackathon survey, we requested contributors which hackathon classes
they have been enthusiastic about attending. Extra folks attended every session
than we anticipated, which indicated that the introduction session on
Monday was crucial for spurring curiosity.
In line with our survey, contributors walked away with a larger understanding of Widespread Fund databases and instruments, so we achieved our foremost purpose of accelerating familiarity with the various datasets supported by the Widespread Fund Information Ecosystem. Moreover, our crew of trainers recognized new Widespread Fund datasets that we plan on integrating into our coaching program sooner or later.
A standard remark was that some classes felt extra like webinars or
workshops than what the title “hackathon” implies. For our subsequent
hackathon, we’ll work with presenters to outline classes as webinars
(an illustration of a knowledge software), workshops (a coaching occasion with
stay coding), or hackathons (an outlined downside that contributors work
on). We additionally obtained requests for extra superior discover and
details about the content material of classes, which we’ll incorporate
into our subsequent spherical of occasion planning.
This occasion, together with many different on-line occasions, lacked the sense of
neighborhood that may be current with in particular person multi-day occasions. We
tried utilizing GitHub Points or Discussions to foster conversations
between contributors, however these instruments have been not often used. We’re
serious about methods to deal with this for our subsequent occasion, and we’re open
to suggestions!
Lastly, the hackathon coordination crew wish to reiterate our
because of all Widespread Fund teams that ran classes for this occasion! We
couldn’t have achieved a variety of datasets and instruments at this
occasion with out your time and efforts.
Subsequent steps
We’re excited to announce that the second CFDE Hackathon will take
place April 25 – twenty ninth! We will positive tune the occasion
with the suggestions from our February occasion, and we hope you’ll be a part of
us!
If you’re enthusiastic about studying extra about attending the April 2022
Hackathon as a participant, please
register right here!
We hope to see you there 🙂
The Widespread Fund Information Ecosystem Coaching Program is funded by the
Nationwide Institutes of Well being (1OT3OD025459-01).
