Di Xiao

Research officer | Computational Systems Biology Group & ProCan | Children's Medical Research Institute | The University of Sydney

[Zhaoxiang (Simon) CAI @zhaoxiang_cai]

Super excited to share our new paper introducing MOSA, a generative model that integrates and augments 7 different types of cancer data across 1,500+ cell lines! 🧬🔬 https://t.co/rPfiv3VEXd @Garnettlab
@sangerinstitute
@CMRI_AUS
@InescID
@istecnico

[Pragathi Masamsetti @PMasamsetti]

New preprint from Tam Lab @CMRI_AUS
- In this study, we investigated and characterized the mesendoderm progenitor in invitro differentiation models of mouse, human embryonic stem cells and in gastrulating mouse embryo.

[Carissa Chen @carissaynchen]

Our benchmarking of Spatially Variable Gene (SVG) detection methods is finally out in @GenomeBiology! Thank you to @hani_jieun and my PhD supervisor @PengyiYang82, as well as @CMRI_AUS @sydneybioinfo for their support!
See the🧵 for more details! https://t.co/EcrbETgEBz

[Computational Systems Biology Lab @PengyiYang82]

Look for a "wet-lab" postdoc researcher to join our ComputationalSystemsBiology lab at Children's Medical Research Institute. Send your app if you work with stem cells @AusStemCell @StemCellsAus @the_ASSCR and epigenetics @AEpiA. Ads:
https://t.co/y5kt862xhF. Close date: 10 Jan

[Katie Zyner @katie_zyner]

🚨Job Alert! Interested in the 🧬#epigenome and #cell_fate_decisions? We are looking for a highly motivated #postdoc to join our multi-disciplinary lab at @CMRI_AUS 🇦🇺. See below for more details. Please RT

https://t.co/1OWjVZJR6R

[Computational Systems Biology Lab @PengyiYang82]

Our recent study (https://t.co/ldLrTzlS6g) highlighted the potential use of deep learning for selecting features in scRNA-seq data. Our new lab study (https://t.co/56afmwHs7f) by @haohuang1999, @ChunleiLiu0, and @manojmwagle examines their suitability and necessary improvements.

[Computational Systems Biology Lab @PengyiYang82]

I'm extremely proud to see our work on ensemble deep learning applied to kinase-substrate prediction using phosphoproteomics data published https://t.co/VoyvuNVgXi. This wouldn't be possible without @SeanJHumphrey @Ben_Leo_Parker @jimmib_78 and outstanding PhD student @DiXiao22.

[Di Xiao @DiXiao22]

Last but not least, thanks to our amazing reviewers and editors who helped substantially improve the quality of this work!

[Di Xiao @DiXiao22]

Thrilled to share our latest publication in #NAR_Genomics_and_Bioinformatics – SnapKin, a cutting-edge ensemble deep learning method for kinase-substrate prediction, that's been selected as Editor’s choice 🌟! Dive into our findings here: https://t.co/yKUJG6L07Z

[Di Xiao @DiXiao22]

7/7 This work would not be possible without my wonderful colleagues at @sydneybioinfo @CMRI_AUS In particular, the co-authors Michael Lin, Chunlei Liu, Thomas A Geddes, our wonderful collaborators @SeanJHumphrey @Ben_Leo_Parker @jimmib_78, and my amazing supervisor @PengyiYang82.

[Di Xiao @DiXiao22]

4/7 SnapKin's superior predictive capabilities are evident as it consistently ranks known kinase substrates higher than alternate methods. This performance establishes it as a vital tool in the phosphoproteome research toolkit.

[Di Xiao @DiXiao22]

3/7 Standing out in a competitive field, SnapKin integrates both 'pseudo-positive' and ensemble strategies within a snapshot model, and leverages CKSAAP encoding to refine predictions, setting a new benchmark for kinase-substrate prediction models in phosphoproteomic research.