Sourya Bhattacharyya Profile
Sourya Bhattacharyya

@SouryaBhattach5

Followers
43
Following
35
Media
0
Statuses
147

Lead Data Scientist, Empirico, San Diego, USA Postdoc + Instructor at La Jolla Institute for Immunology Ph.D. Computer Science at IIT Kharagpur

San Diego
Joined September 2018
Don't wanna be here? Send us removal request.
@SouryaBhattach5
Sourya Bhattacharyya
3 months
RT @BiologyAIDaily: Variant effect predictor correlation with functional assays is reflective of clinical classification performance. 1. Th….
0
2
0
@SouryaBhattach5
Sourya Bhattacharyya
4 months
RT @tangming2005: The epigenetic hallmarks of immune cells in cancer
Tweet media one
0
17
0
@SouryaBhattach5
Sourya Bhattacharyya
6 months
RT @ferhatay: Happy to share Dante Bolzan’s delicious new preprint-Soffritto. He developed a deep learning model that accurately predicts h….
0
3
0
@SouryaBhattach5
Sourya Bhattacharyya
6 months
Incredibly thankful to my post doc supervisor Dr. @ferhatay for all his guidance and comments in shaping up this paper, which should serve as a great tool to the community.
0
0
0
@SouryaBhattach5
Sourya Bhattacharyya
6 months
It was wonderful to mentor Daniela Salgado Figueroa, Ph.D. student in UCSD, who did an incredible work on validating DiffHiChIP.
0
0
0
@SouryaBhattach5
Sourya Bhattacharyya
6 months
We have tested with various HiChIP datasets between two conditions, with small (=2) or large (=6) number of replicates, confirming the utility of DiffHiChIP as a go-to model for calling differential HiChIP (and other 3C) loops.
0
0
0
@SouryaBhattach5
Sourya Bhattacharyya
6 months
DiffHiChIP is also benchmarked with compatible Hi-C loops, and various experimentally validated reference chromatin contacts from validation experiments.
0
0
0
@SouryaBhattach5
Sourya Bhattacharyya
6 months
DiffHiChIP shows that generalized linear models (GLM) in edgeR employing either likelihood ratio test (LRT) or quasi-likelihood F test, coupled with distance stratification via independent hypothesis weighting (IHW), shows better detection of differential long-range HiChIP loops.
0
0
0
@SouryaBhattach5
Sourya Bhattacharyya
6 months
Extremely happy to share that our work DiffHiChIP: a method to identify differential chromatin interactions from HiChIP (and any other chromatin interaction data such as Hi-C, ChIA-PET), is now online. Preprint in:
www.biorxiv.org
High-resolution conformation capture assays such as HiChIP are commonly used for profiling chromatin loops among cis -regulatory elements including enhancers and promoters. Detection of differential...
5
3
8
@SouryaBhattach5
Sourya Bhattacharyya
9 months
I will be @ASHG2024, Denver, CO, Nov 5-9 along with my colleagues at @Empirico (. Looking forward to discuss latest research in computational and ML models for Human Genetics. Let me know if anybody wants to catch up.
www.empiri.co
Empirico combines human genetics-driven target discovery with siRNA drug discovery, across a range of common diseases with unmet medical need.
1
0
1
@SouryaBhattach5
Sourya Bhattacharyya
10 months
RT @pkoo562: Excited that CREME (Cis-Regulatory Element Model Explanations) is published in @NatureGenet! Work led by @ToneyanSh!. tl;dr CR….
0
43
0
@SouryaBhattach5
Sourya Bhattacharyya
1 year
RT @ferhatay: Forgot a little detail, the link itself :) let us know if you have any feedback:
0
1
0
@SouryaBhattach5
Sourya Bhattacharyya
1 year
RT @ferhatay: Our multi-year effort on putting together a comprehensive database of loop calls from 1000+ HiChIP samples plus enabling visu….
0
13
0
@SouryaBhattach5
Sourya Bhattacharyya
1 year
Looking forward to attend the CSHL Genome Organization & Nuclear Function meeting @cshlmeetings from April 30 - May 4. Hope to listen great lectures, meet and discuss with the stalwarts of 3D genome research.
0
1
3
@SouryaBhattach5
Sourya Bhattacharyya
1 year
RT @biorxiv_bioinfo: A deep profile of gene expression across 18 human cancers #biorxiv_bioinfo.
0
1
0
@SouryaBhattach5
Sourya Bhattacharyya
1 year
RT @LB_Barreiro: Finally, our data indicates that epigenetic QTLs provide a means to markedly increase the number of QTLs that colocalize (….
0
4
0