Cross-platform representation and visualization of biological data in flutter.

Features

Widgets:

• UMAP Visualizer:

Biological Data Classes:

• Protein
• Sequence: AminoAcidSequence, NucleotideSequence
• Protein-Protein Interaction
• Taxonomy

Protein Representation and Data Analysis:

• Embedding: PerSequence, PerResidue
• UMAP

File handling:

Usage

Load and write a protein fasta file:

import 'package:flutter_test/flutter_test.dart';

Future<List<Protein>> readFastaFileProtein(String pathToFile) async {
  BioFileHandlerContext<Protein>? handler = BioFileHandler<Protein>().create(pathToFile);
  Map<String, Protein> proteins = await handler.read();
  return proteins.values.toList();
}

Future<void> saveProteins(List<Protein> proteins, String pathToFile) async {
  // Saves the file on desktop/mobile, downloads on web
  BioFileHandlerContext<Protein>? handler = BioFileHandler<Protein>().create(pathToFile);
  await handler.write(proteins.asMap().map((_, value) => MapEntry(value.id, value)));
}

Add custom attributes to your proteins:

import 'package:flutter_test/flutter_test.dart';

Future<List<Protein>> addCustomAttributes(List<Protein> proteins, String pathToFile) async {
  BioFileHandlerContext<CustomAttributes>? handler = BioFileHandler<CustomAttributes>().create(pathToFile);
  Map<String, CustomAttributes> attributes = await handler.read();
  List<Protein> updatedProteins = [];
  for (Protein protein in proteins) {
    if (attributes.containsKey(protein.id)) {
      Protein updated = protein.updateFromCustomAttributes(attributes[protein.id]!);
      updatedProteins.add(updated);
    }
  }
  return updatedProteins;
}

Create a UMAP Visualizer Widget with random coordinates:

import 'package:flutter_test/flutter_test.dart';

Widget createUmapWidget(List<ProteinProteinInteraction> interactionData) {
  return UmapVisualizer(
      umapData: UMAPData.random(interactionData.length),
      pointIdentifierKey: "id",
      pointData: interactionData.map((interaction) => interaction.toMap()).toList()); // Also works with protein data
}

Additional information

The correctness of the provided operations is the most important focus of this package in order to be useful for scientists and practitioners in research and development. We are grateful for every reported bug and contribution to the codebase on GitHub!

Current Roadmap:

• Increase test coverage
• Improve documentation
• Support more file formats for all data types
• Add 3D protein visualization