dimensions.py

import requests
import pandas as pd
import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt
from matplotlib.colors import LogNorm

URL = 'https://app.dimensions.ai/viz/data/publication/timeline-source-published.json?search_mode=content&search_text="{engine}" AND "proteomics"&search_type=kws&search_field=full_search'

def get(engine: str):
    data = requests.get(URL.format(engine=engine)).json()
    pubs = np.array(data['data']).flatten()
    years = [x['id'] for x in data['dimensions'][1]['items']]
    print(pubs)
    print(years)
    df = pd.DataFrame(data=pubs, index=years)
    df['engine'] = engine
    return df

free = [
    "Comet",
    "Morpheus",
    "X!Tandem",
    "OMSSA",
    "MyriMatch",
    "IdentiPy"
]

prop = [
    "MaxQuant",
    "Proteome Discoverer",
    "Mascot",
    "Sequest",
    "ProteinPilot",
    "Byonic",
    "MSFragger",
    "PEAKS DB",
    "MS-GF",
    "MS Amanda",
    "Spectrum Mill",
    "pFind",
    "CHIMERYS"
]

dfs = []
for engine in free:
    df = get(engine)
    df["free"] = True
    dfs.append(df)
for engine in prop:
    df = get(engine)
    df["free"] = False
    dfs.append(df)

df = pd.concat(dfs).reset_index().rename({0: 'pubs', 'index': 'year'},axis=1)
df = df[df.year < 2023]

fig, (ax0, ax1) = plt.subplots(nrows=1,ncols=2)
t = df.pivot_table(index='year', columns='engine', values='pubs').T.sort_values(by=2022, ascending=False)
sns.heatmap(t, annot=True, fmt='d', norm=LogNorm(), cmap='mako', ax=ax0)

p = df.pivot_table(index='year', columns='free', values='pubs', aggfunc='sum')
ax1.plot(p.index, p[False], label='Closed source or non-free')
ax1.plot(p.index, p[True], label='Open source and free')
ax1.legend()
fig.suptitle("Publications mentioning DDA search engine (dimensions.ai)")
plt.show()