Build an enrichment tool analyser in Java.

Develop a tool to determine the genetic background of a given sequencing library from mouse origin; for this, the tool will rely on the comprehensive mouse strain information collected by the Mouse Genomes Project. The aim is to identify at an early stage whether a sample really is derived from the mouse line or cross the researcher thinks it is… Tentative name: reStrainingOrder

There a different protocols to generate bisulfite sequencing libraries, and depending on how the library was made, different processing of the sequencing data is required. Charades is a tool that aims to determine the nature of a bisulfite sequencing library from the base composition in the corresponding fastq file. I started on Charades at the last hackathon but unfortunately, the project has been on ice for a bit - it's time to pick it up again!

I'm going to work on completing a project from a previous hackathon which is a gene ontology search tool which incorporates knoweldge of known sources of artefacts which can affect the validity of such search results, and will try to identify and highlight these.

I'll be working on a gene ontology tool from a previous hackathon, to add functionality for identifying potential biases and artefacts in the results. 

Neural networks for topologically associated domain (TAD) boundary detection 

I will be continuing work on implementing a deep convolutional neural network for the detection of the genomic location of TAD boundaries. The implementation will be done using tensor flow and the network will be trained on publicly available data. As a control, I have Hi-C data from cells under conditions where no TAD organization is present. This approach should allow for straight-forward integration of replicate data. The problem is also parallelizable so an implementation using GPGPU (or other massively parallel processing) can be done, although this is likely beyond the scope of this hackathon.

For a more detailed explanation about TADs and this project, please visit the repo

I plan to complete ParticleStats2.0 with new features for whole organism behavioural assays and updates for tools for comparing tracked particles between experiments. I worked on this project at last years hackathon and it is approaching completion.

I'm going to be working on developing a PCA based QC tool to be implemented as part of our RNA-Seq pipeline

Protein secondary structure prediction using Deep Neural Networks

Deep learning methods for matrix classification problems in high-throughput screening.

Predicting the cell types of single-cell RNA-Seq samples

Single-cell RNA-Seq makes it possible to characterize the transcriptomes of cell types and identify their transcriptional signatures via differential analysis. The challenge is to create a machine learning model for identifying cell types from a dataset of single-cell RNA samples. You can use the method described here:

https://www.biorxiv.org/content/biorxiv/early/2018/02/14/258566.full.pdf

People would be more than welcome to join our team.